Search Results for author: Mohit Bansal

Found 322 papers, 209 papers with code

Multimodal Intent Discovery from Livestream Videos

no code implementations Findings (NAACL) 2022 Adyasha Maharana, Quan Tran, Franck Dernoncourt, Seunghyun Yoon, Trung Bui, Walter Chang, Mohit Bansal

We construct and present a new multimodal dataset consisting of software instructional livestreams and containing manual annotations for both detailed and abstract procedural intent that enable training and evaluation of joint video and text understanding models.

Intent Discovery Video Summarization +1

Continual Few-Shot Learning for Text Classification

1 code implementation EMNLP 2021 Ramakanth Pasunuru, Veselin Stoyanov, Mohit Bansal

In this work, we propose a continual few-shot learning (CFL) task, in which a system is challenged with a difficult phenomenon and asked to learn to correct mistakes with only a few (10 to 15) training examples.

continual few-shot learning Few-Shot Learning +4

An Overview of Uncertainty Calibration for Text Classification and the Role of Distillation

no code implementations ACL (RepL4NLP) 2021 Han Guo, Ramakanth Pasunuru, Mohit Bansal

Many recalibration methods have been proposed in the literature for quantifying predictive uncertainty and calibrating model outputs, with varying degrees of complexity.

text-classification Text Classification

Integrating Visuospatial, Linguistic, and Commonsense Structure into Story Visualization

1 code implementation EMNLP 2021 Adyasha Maharana, Mohit Bansal

Such information is even more important for story visualization since its inputs have an explicit narrative structure that needs to be translated into an image sequence (or visual story).

Dense Captioning Image Generation +1

NDH-Full: Learning and Evaluating Navigational Agents on Full-Length Dialogue

1 code implementation EMNLP 2021 Hyounghun Kim, Jialu Li, Mohit Bansal

In this paper, we explore the Navigation from Dialogue History (NDH) task, which is based on the Cooperative Vision-and-Dialogue Navigation (CVDN) dataset, and present a state-of-the-art model which is built upon Vision-Language transformers.

Data Augmentation Dynamic Time Warping +1

Inducing Transformer’s Compositional Generalization Ability via Auxiliary Sequence Prediction Tasks

1 code implementation EMNLP 2021 Yichen Jiang, Mohit Bansal

Motivated by the failure of a Transformer model on the SCAN compositionality challenge (Lake and Baroni, 2018), which requires parsing a command into actions, we propose two auxiliary sequence prediction tasks as additional training supervision.

GraDA: Graph Generative Data Augmentation for Commonsense Reasoning

1 code implementation COLING 2022 Adyasha Maharana, Mohit Bansal

Recent advances in commonsense reasoning have been fueled by the availability of large-scale human annotated datasets.

Data Augmentation HellaSwag +1

On Curriculum Learning for Commonsense Reasoning

1 code implementation NAACL 2022 Adyasha Maharana, Mohit Bansal

Hence, we examine the effect of a human-like easy-to-difficult curriculum during finetuning of language models for commonsense reasoning tasks.

HellaSwag Learning-To-Rank +3

Enhancing Knowledge Selection for Grounded Dialogues via Document Semantic Graphs

no code implementations NAACL 2022 Sha Li, Mahdi Namazifar, Di Jin, Mohit Bansal, Heng Ji, Yang Liu, Dilek Hakkani-Tur

In this work, we propose to automatically convert the background knowledge documents into document semantic graphs and then perform knowledge selection over such graphs.

Multi-Task Learning Response Generation +1

Unlearning Sensitive Information in Multimodal LLMs: Benchmark and Attack-Defense Evaluation

1 code implementation1 May 2025 Vaidehi Patil, Yi-Lin Sung, Peter Hase, Jie Peng, Tianlong Chen, Mohit Bansal

To address this gap, we first introduce a multimodal unlearning benchmark, UnLOK-VQA (Unlearning Outside Knowledge VQA), as well as an attack-and-defense framework to evaluate methods for deleting specific multimodal knowledge from MLLMs.

Question Answering Specificity +1

A Comprehensive Survey in LLM(-Agent) Full Stack Safety: Data, Training and Deployment

no code implementations22 Apr 2025 Kun Wang, Guibin Zhang, Zhenhong Zhou, Jiahao Wu, Miao Yu, Shiqian Zhao, Chenlong Yin, Jinhu Fu, Yibo Yan, Hanjun Luo, Liang Lin, Zhihao Xu, Haolang Lu, Xinye Cao, Xinyun Zhou, Weifei Jin, Fanci Meng, Junyuan Mao, Yu Wang, Hao Wu, Minghe Wang, Fan Zhang, Junfeng Fang, Wenjie Qu, Yue Liu, Chengwei Liu, Yifan Zhang, Qiankun Li, Chongye Guo, Yalan Qin, Zhaoxin Fan, Yi Ding, Donghai Hong, Jiaming Ji, Yingxin Lai, Zitong Yu, Xinfeng Li, Yifan Jiang, Yanhui Li, Xinyu Deng, Junlin Wu, Dongxia Wang, Yihao Huang, Yufei Guo, Jen-tse Huang, Qiufeng Wang, Wenxuan Wang, Dongrui Liu, Yanwei Yue, Wenke Huang, Guancheng Wan, Heng Chang, Tianlin Li, Yi Yu, Chenghao Li, Jiawei Li, Lei Bai, Jie Zhang, Qing Guo, Jingyi Wang, Tianlong Chen, Joey Tianyi Zhou, Xiaojun Jia, Weisong Sun, Cong Wu, Jing Chen, Xuming Hu, Yiming Li, Xiao Wang, Ningyu Zhang, Luu Anh Tuan, Guowen Xu, Jiaheng Zhang, Tianwei Zhang, Xingjun Ma, Jindong Gu, Xiang Wang, Bo An, Jun Sun, Mohit Bansal, Shirui Pan, Lingjuan Lyu, Yuval Elovici, Bhavya Kailkhura, Yaodong Yang, Hongwei Li, Wenyuan Xu, Yizhou Sun, Wei Wang, Qing Li, Ke Tang, Yu-Gang Jiang, Felix Juefei-Xu, Hui Xiong, XiaoFeng Wang, DaCheng Tao, Philip S. Yu, Qingsong Wen, Yang Liu

Currently, existing surveys on LLM safety primarily focus on specific stages of the LLM lifecycle, e. g., deployment phase or fine-tuning phase, lacking a comprehensive understanding of the entire "lifechain" of LLMs.

Model Editing

Retrieval-Augmented Generation with Conflicting Evidence

1 code implementation17 Apr 2025 Han Wang, Archiki Prasad, Elias Stengel-Eskin, Mohit Bansal

Large language model (LLM) agents are increasingly employing retrieval-augmented generation (RAG) to improve the factuality of their responses.

Misinformation RAG +2

Training-free Guidance in Text-to-Video Generation via Multimodal Planning and Structured Noise Initialization

no code implementations11 Apr 2025 Jialu Li, Shoubin Yu, Han Lin, Jaemin Cho, Jaehong Yoon, Mohit Bansal

Video-MSG consists of three steps, where in the first two steps, Video-MSG creates Video Sketch, a fine-grained spatio-temporal plan for the final video, specifying background, foreground, and object trajectories, in the form of draft video frames.

Denoising Object +3

Task-Circuit Quantization: Leveraging Knowledge Localization and Interpretability for Compression

1 code implementation10 Apr 2025 Hanqi Xiao, Yi-Lin Sung, Elias Stengel-Eskin, Mohit Bansal

We compare TaCQ-based quantization to existing mixed-precision quantization methods when conditioning both on general-purpose and task-specific data.

Math MMLU +2

CoKe: Customizable Fine-Grained Story Evaluation via Chain-of-Keyword Rationalization

no code implementations21 Mar 2025 Brihi Joshi, Sriram Venkatapathy, Mohit Bansal, Nanyun Peng, Haw-Shiuan Chang

Evaluating creative text such as human-written stories using language models has always been a challenging task -- owing to the subjectivity of multi-annotator ratings.

MAMM-Refine: A Recipe for Improving Faithfulness in Generation with Multi-Agent Collaboration

no code implementations19 Mar 2025 David Wan, Justin Chih-Yao Chen, Elias Stengel-Eskin, Mohit Bansal

We investigate how iterative collaboration among multiple instances and types of large language models (LLMs) enhances subtasks in the refinement process, such as error detection, critiquing unfaithful sentences, and making corrections based on critiques.

Long Form Question Answering Reranking

VEGGIE: Instructional Editing and Reasoning Video Concepts with Grounded Generation

no code implementations18 Mar 2025 Shoubin Yu, Difan Liu, Ziqiao Ma, Yicong Hong, Yang Zhou, Hao Tan, Joyce Chai, Mohit Bansal

To support diverse tasks and complex instructions, we employ a curriculum learning strategy: first aligning the MLLM and video diffusion model with large-scale instructional image editing data, followed by end-to-end fine-tuning on high-quality multitask video data.

Reasoning Segmentation Video Editing

Symbolic Mixture-of-Experts: Adaptive Skill-based Routing for Heterogeneous Reasoning

no code implementations7 Mar 2025 Justin Chih-Yao Chen, Sukwon Yun, Elias Stengel-Eskin, Tianlong Chen, Mohit Bansal

We propose a skill-based recruiting strategy that dynamically selects the most relevant set of expert LLMs for diverse reasoning tasks based on their strengths.

Math Mixture-of-Experts +1

RSQ: Learning from Important Tokens Leads to Better Quantized LLMs

1 code implementation3 Mar 2025 Yi-Lin Sung, Prateek Yadav, Jialu Li, Jaehong Yoon, Mohit Bansal

Building on this finding, we propose RSQ (Rotate, Scale, then Quantize), which (1) applies rotations (orthogonal transformation) to the model to mitigate outliers (those with exceptionally large magnitude), (2) scales the token feature based on its importance, and (3) quantizes the model using the GPTQ framework with the second-order statistics computed by scaled tokens.

Quantization

MutaGReP: Execution-Free Repository-Grounded Plan Search for Code-Use

no code implementations21 Feb 2025 Zaid Khan, Ali Farhadi, Ranjay Krishna, Luca Weihs, Mohit Bansal, Tanmay Gupta

When a human requests an LLM to complete a coding task using functionality from a large code repository, how do we provide context from the repo to the LLM?

Navigate

UPCORE: Utility-Preserving Coreset Selection for Balanced Unlearning

1 code implementation20 Feb 2025 Vaidehi Patil, Elias Stengel-Eskin, Mohit Bansal

We find that UPCORE improves both standard metrics and AUC, benefitting from positive transfer between the coreset and pruned points while reducing negative transfer from the forget set to points outside of it.

Learning to Generate Unit Tests for Automated Debugging

1 code implementation3 Feb 2025 Archiki Prasad, Elias Stengel-Eskin, Justin Chih-Yao Chen, Zaid Khan, Mohit Bansal

However, we uncover a trade-off between generating unit test inputs that reveal errors when given a faulty code and correctly predicting the unit test output without access to the gold solution.

HumanEval Large Language Model +1

TimeRefine: Temporal Grounding with Time Refining Video LLM

1 code implementation12 Dec 2024 Xizi Wang, Feng Cheng, Ziyang Wang, Huiyu Wang, Md Mohaiminul Islam, Lorenzo Torresani, Mohit Bansal, Gedas Bertasius, David Crandall

Recent work has focused on enabling Video LLMs to perform video temporal grounding via next-token prediction of temporal timestamps.

Temporal Localization

Bootstrapping Language-Guided Navigation Learning with Self-Refining Data Flywheel

1 code implementation11 Dec 2024 Zun Wang, Jialu Li, Yicong Hong, Songze Li, Kunchang Li, Shoubin Yu, Yi Wang, Yu Qiao, Yali Wang, Mohit Bansal, LiMin Wang

In this paper, we introduce a Self-Refining Data Flywheel (SRDF) that generates high-quality and large-scale navigational instruction-trajectory pairs by iteratively refining the data pool through the collaboration between two models, the instruction generator and the navigator, without any human-in-the-loop annotation.

QAPyramid: Fine-grained Evaluation of Content Selection for Text Summarization

1 code implementation10 Dec 2024 Shiyue Zhang, David Wan, Arie Cattan, Ayal Klein, Ido Dagan, Mohit Bansal

The Pyramid human evaluation protocol, which assesses content selection by breaking the reference summary into sub-units and verifying their presence in the system summary, has been widely adopted.

Text Summarization

SAME: Learning Generic Language-Guided Visual Navigation with State-Adaptive Mixture of Experts

1 code implementation7 Dec 2024 Gengze Zhou, Yicong Hong, Zun Wang, Chongyang Zhao, Mohit Bansal, Qi Wu

The academic field of learning instruction-guided visual navigation can be generally categorized into high-level category-specific search and low-level language-guided navigation, depending on the granularity of language instruction, in which the former emphasizes the exploration process, while the latter concentrates on following detailed textual commands.

General Knowledge Mixture-of-Experts +1

DreamRunner: Fine-Grained Storytelling Video Generation with Retrieval-Augmented Motion Adaptation

no code implementations25 Nov 2024 Zun Wang, Jialu Li, Han Lin, Jaehong Yoon, Mohit Bansal

To address these challenges, we propose DreamRunner, a novel story-to-video generation method: First, we structure the input script using a large language model (LLM) to facilitate both coarse-grained scene planning as well as fine-grained object-level layout and motion planning.

Large Language Model Motion Planning +4

VideoRepair: Improving Text-to-Video Generation via Misalignment Evaluation and Localized Refinement

no code implementations22 Nov 2024 Daeun Lee, Jaehong Yoon, Jaemin Cho, Mohit Bansal

In (2) refinement planning, we identify accurately generated objects and then create localized prompts to refine other areas in the video.

Text-to-Video Generation Video Alignment +1

Motion-Grounded Video Reasoning: Understanding and Perceiving Motion at Pixel Level

no code implementations15 Nov 2024 Andong Deng, Tongjia Chen, Shoubin Yu, Taojiannan Yang, Lincoln Spencer, Yapeng Tian, Ajmal Saeed Mian, Mohit Bansal, Chen Chen

In this paper, we introduce Motion-Grounded Video Reasoning, a new motion understanding task that requires generating visual answers (video segmentation masks) according to the input question, and hence needs implicit spatiotemporal reasoning and grounding.

Benchmarking counterfactual +6

M3DocRAG: Multi-modal Retrieval is What You Need for Multi-page Multi-document Understanding

no code implementations7 Nov 2024 Jaemin Cho, Debanjan Mahata, Ozan Irsoy, Yujie He, Mohit Bansal

However, there are difficulties in applying these methods in real-world scenarios: (a) questions often require information across different pages or documents, where MLMs cannot handle many long documents; (b) documents often have important information in visual elements such as figures, but text extraction tools ignore them.

document understanding Optical Character Recognition +6

Self-Consistency Preference Optimization

no code implementations6 Nov 2024 Archiki Prasad, Weizhe Yuan, Richard Yuanzhe Pang, Jing Xu, Maryam Fazel-Zarandi, Mohit Bansal, Sainbayar Sukhbaatar, Jason Weston, Jane Yu

Self-alignment, whereby models learn to improve themselves without human annotation, is a rapidly growing research area.

GSM8K Math

Explaining and Improving Contrastive Decoding by Extrapolating the Probabilities of a Huge and Hypothetical LM

1 code implementation3 Nov 2024 Haw-Shiuan Chang, Nanyun Peng, Mohit Bansal, Anil Ramakrishna, Tagyoung Chung

In FactualityPrompts, an open-ended text generation benchmark, sampling using APD significantly boosts factuality in comparison to the CD sampling and its variants, and achieves state-of-the-art results for Pythia 6. 9B and OPT 6. 7B.

LAMBADA Text Generation

On Positional Bias of Faithfulness for Long-form Summarization

1 code implementation31 Oct 2024 David Wan, Jesse Vig, Mohit Bansal, Shafiq Joty

Large Language Models (LLMs) often exhibit positional bias in long-context settings, under-attending to information in the middle of inputs.

Form

Unbounded: A Generative Infinite Game of Character Life Simulation

no code implementations24 Oct 2024 Jialu Li, Yuanzhen Li, Neal Wadhwa, Yael Pritch, David E. Jacobs, Michael Rubinstein, Mohit Bansal, Nataniel Ruiz

We introduce the concept of a generative infinite game, a video game that transcends the traditional boundaries of finite, hard-coded systems by using generative models.

Instruction Following Language Modelling +1

Teaching Models to Balance Resisting and Accepting Persuasion

1 code implementation18 Oct 2024 Elias Stengel-Eskin, Peter Hase, Mohit Bansal

PBT consistently improves resistance to misinformation and resilience to being challenged while also resulting in the best overall performance on holistic data containing both positive and negative persuasion.

Misinformation

SAFREE: Training-Free and Adaptive Guard for Safe Text-to-Image And Video Generation

1 code implementation16 Oct 2024 Jaehong Yoon, Shoubin Yu, Vaidehi Patil, Huaxiu Yao, Mohit Bansal

To address these, we propose SAFREE, a novel, training-free approach for safe T2I and T2V, that does not alter the model's weights.

Denoising Video Generation

Adapt-$\infty$: Scalable Lifelong Multimodal Instruction Tuning via Dynamic Data Selection

1 code implementation14 Oct 2024 Adyasha Maharana, Jaehong Yoon, Tianlong Chen, Mohit Bansal

This data selector samples a subset of the most important samples from each skill cluster for training.

LLM Self-Correction with DeCRIM: Decompose, Critique, and Refine for Enhanced Following of Instructions with Multiple Constraints

no code implementations9 Oct 2024 Thomas Palmeira Ferraz, Kartik Mehta, Yu-Hsiang Lin, Haw-Shiuan Chang, Shereen Oraby, Sijia Liu, Vivek Subramanian, Tagyoung Chung, Mohit Bansal, Nanyun Peng

To address the performance gap between open-source and proprietary models, we propose the Decompose, Critique and Refine (DeCRIM) self-correction pipeline, which enhances LLMs' ability to follow constraints.

Instruction Following

Glider: Global and Local Instruction-Driven Expert Router

1 code implementation9 Oct 2024 Pingzhi Li, Prateek Yadav, Jaehong Yoon, Jie Peng, Yi-Lin Sung, Mohit Bansal, Tianlong Chen

Our experiments using T5-based models for T0 and FLAN tasks demonstrate that GLIDER achieves substantially improved held-in performance while maintaining strong generalization on held-out tasks.

Localizing Factual Inconsistencies in Attributable Text Generation

1 code implementation9 Oct 2024 Arie Cattan, Paul Roit, Shiyue Zhang, David Wan, Roee Aharoni, Idan Szpektor, Mohit Bansal, Ido Dagan

There has been an increasing interest in detecting hallucinations in model-generated texts, both manually and automatically, at varying levels of granularity.

Text Generation

DataEnvGym: Data Generation Agents in Teacher Environments with Student Feedback

1 code implementation8 Oct 2024 Zaid Khan, Elias Stengel-Eskin, Jaemin Cho, Mohit Bansal

Students are iteratively trained and evaluated on generated data, and their feedback (in the form of errors or weak skills) is reported to the agent after each iteration.

Math Sequential Decision Making +1

What Matters for Model Merging at Scale?

no code implementations4 Oct 2024 Prateek Yadav, Tu Vu, Jonathan Lai, Alexandra Chronopoulou, Manaal Faruqui, Mohit Bansal, Tsendsuren Munkhdalai

This leaves many unanswered questions about the effect of scaling model size and how it interplays with other key factors -- like the base model quality and number of expert models -- , to affect the merged model's performance.

model Task Arithmetic +1

VEDIT: Latent Prediction Architecture For Procedural Video Representation Learning

no code implementations4 Oct 2024 Han Lin, Tushar Nagarajan, Nicolas Ballas, Mido Assran, Mojtaba Komeili, Mohit Bansal, Koustuv Sinha

In this work, we show that a strong off-the-shelf frozen pretrained visual encoder, along with a well designed prediction model, can achieve state-of-the-art (SoTA) performance in forecasting and procedural planning without the need for pretraining the prediction model, nor requiring additional supervision from language or ASR.

Action Anticipation Denoising +2

LASeR: Learning to Adaptively Select Reward Models with Multi-Armed Bandits

1 code implementation2 Oct 2024 Duy Nguyen, Archiki Prasad, Elias Stengel-Eskin, Mohit Bansal

Our results on commonsense and math reasoning tasks demonstrate that LASeR can boost iterative LLM optimization by optimizing for multiple RMs, improving the absolute average accuracy of Llama-3-8B over three datasets by 2. 67% over training with ensemble RM scores while also showing superior training efficiency (e. g., a 2x speedup).

Instruction Following Math +1

MAgICoRe: Multi-Agent, Iterative, Coarse-to-Fine Refinement for Reasoning

1 code implementation18 Sep 2024 Justin Chih-Yao Chen, Archiki Prasad, Swarnadeep Saha, Elias Stengel-Eskin, Mohit Bansal

Moreover, to ensure effective refinement, we employ a multi-agent loop with three agents: Solver, Reviewer (which generates targeted feedback based on step-wise RM scores), and the Refiner (which incorporates feedback).

Math

AdaCAD: Adaptively Decoding to Balance Conflicts between Contextual and Parametric Knowledge

1 code implementation11 Sep 2024 Han Wang, Archiki Prasad, Elias Stengel-Eskin, Mohit Bansal

Knowledge conflict arises from discrepancies between information in the context of a large language model (LLM) and the knowledge stored in its parameters.

Language Modelling Large Language Model +1

A Survey on Model MoErging: Recycling and Routing Among Specialized Experts for Collaborative Learning

no code implementations13 Aug 2024 Prateek Yadav, Colin Raffel, Mohammed Muqeeth, Lucas Caccia, Haokun Liu, Tianlong Chen, Mohit Bansal, Leshem Choshen, Alessandro Sordoni

The availability of performant pre-trained models has led to a proliferation of fine-tuned expert models that are specialized to a particular domain or task.

Mixture-of-Experts Survey

System-1.x: Learning to Balance Fast and Slow Planning with Language Models

1 code implementation19 Jul 2024 Swarnadeep Saha, Archiki Prasad, Justin Chih-Yao Chen, Peter Hase, Elias Stengel-Eskin, Mohit Bansal

To this end, we propose the System-1. x Planner, a controllable planning framework with LLMs that is capable of generating hybrid plans and balancing between the two planning modes based on the difficulty of the problem at hand.

Vision-and-Language Navigation Today and Tomorrow: A Survey in the Era of Foundation Models

1 code implementation9 Jul 2024 Yue Zhang, Ziqiao Ma, Jialu Li, Yanyuan Qiao, Zun Wang, Joyce Chai, Qi Wu, Mohit Bansal, Parisa Kordjamshidi

Vision-and-Language Navigation (VLN) has gained increasing attention over recent years and many approaches have emerged to advance their development.

Vision and Language Navigation

Fundamental Problems With Model Editing: How Should Rational Belief Revision Work in LLMs?

no code implementations27 Jun 2024 Peter Hase, Thomas Hofweber, Xiang Zhou, Elias Stengel-Eskin, Mohit Bansal

With this goal in mind, this paper critiques the standard formulation of the model editing problem and proposes a formal testbed for model editing research.

Model Editing Philosophy

See It from My Perspective: Diagnosing the Western Cultural Bias of Large Vision-Language Models in Image Understanding

1 code implementation17 Jun 2024 Amith Ananthram, Elias Stengel-Eskin, Carl Vondrick, Mohit Bansal, Kathleen McKeown

Moreover, while prompting in the language of a target culture can lead to reductions in bias, it is not a substitute for building AI more representative of the world's languages.

REAL Sampling: Boosting Factuality and Diversity of Open-Ended Generation via Asymptotic Entropy

1 code implementation11 Jun 2024 Haw-Shiuan Chang, Nanyun Peng, Mohit Bansal, Anil Ramakrishna, Tagyoung Chung

If a LLM's entropy is higher than the asymptotic entropy (i. e., the LLM is more uncertain than it should be), the THF model predicts a high hallucination hazard, which leads to a lower p threshold in REAL sampling.

Diversity Hallucination

Are language models rational? The case of coherence norms and belief revision

no code implementations5 Jun 2024 Thomas Hofweber, Peter Hase, Elias Stengel-Eskin, Mohit Bansal

We consider both logical coherence norms as well as coherence norms tied to the strength of belief.

The Power of Summary-Source Alignments

1 code implementation2 Jun 2024 Ori Ernst, Ori Shapira, Aviv Slobodkin, Sharon Adar, Mohit Bansal, Jacob Goldberger, Ran Levy, Ido Dagan

Multi-document summarization (MDS) is a challenging task, often decomposed to subtasks of salience and redundancy detection, followed by text generation.

Document Summarization Multi-Document Summarization +1

LACIE: Listener-Aware Finetuning for Confidence Calibration in Large Language Models

2 code implementations31 May 2024 Elias Stengel-Eskin, Peter Hase, Mohit Bansal

To calibrate both implicit and explicit confidence markers, we introduce a pragmatic, listener-aware finetuning method (LACIE) that models the listener, considering not only whether an answer is right, but whether it will be accepted by a listener.

TriviaQA TruthfulQA

VideoTree: Adaptive Tree-based Video Representation for LLM Reasoning on Long Videos

1 code implementation29 May 2024 Ziyang Wang, Shoubin Yu, Elias Stengel-Eskin, Jaehong Yoon, Feng Cheng, Gedas Bertasius, Mohit Bansal

Specifically, we incorporate multigranularity information into a tree-based representation, allowing VideoTree to extract query-relevant details from long videos in a coarse-to-fine manner.

EgoSchema MME +3

RACCooN: A Versatile Instructional Video Editing Framework with Auto-Generated Narratives

2 code implementations28 May 2024 Jaehong Yoon, Shoubin Yu, Mohit Bansal

(3) RACCooN also plans to imagine new objects in a given video, so users simply prompt the model to receive a detailed video editing plan for complex video editing.

Attribute Video Editing

FlexEControl: Flexible and Efficient Multimodal Control for Text-to-Image Generation

no code implementations8 May 2024 Xuehai He, Jian Zheng, Jacob Zhiyuan Fang, Robinson Piramuthu, Mohit Bansal, Vicente Ordonez, Gunnar A Sigurdsson, Nanyun Peng, Xin Eric Wang

Controllable text-to-image (T2I) diffusion models generate images conditioned on both text prompts and semantic inputs of other modalities like edge maps.

Text-to-Image Generation

Ctrl-Adapter: An Efficient and Versatile Framework for Adapting Diverse Controls to Any Diffusion Model

no code implementations15 Apr 2024 Han Lin, Jaemin Cho, Abhay Zala, Mohit Bansal

ControlNets are widely used for adding spatial control to text-to-image diffusion models with different conditions, such as depth maps, scribbles/sketches, and human poses.

Image Generation Style Transfer +3

Rethinking Interactive Image Segmentation with Low Latency, High Quality, and Diverse Prompts

1 code implementation31 Mar 2024 Qin Liu, Jaemin Cho, Mohit Bansal, Marc Niethammer

In light of this, we reintroduce this dense design into the generalist models, to facilitate the development of generalist models with high segmentation quality.

Image Segmentation Interactive Segmentation +2

EnvGen: Generating and Adapting Environments via LLMs for Training Embodied Agents

no code implementations18 Mar 2024 Abhay Zala, Jaemin Cho, Han Lin, Jaehong Yoon, Mohit Bansal

Then, we enable the LLM to continuously adapt the generated environments to progressively improve the skills that the agent is weak at, by providing feedback to the LLM in the form of the agent's performance.

Reinforcement Learning (RL) World Knowledge

DAM: Dynamic Adapter Merging for Continual Video QA Learning

1 code implementation13 Mar 2024 Feng Cheng, Ziyang Wang, Yi-Lin Sung, Yan-Bo Lin, Mohit Bansal, Gedas Bertasius

Our DAM model outperforms prior state-of-the-art continual learning approaches by 9. 1% while exhibiting 1. 9% less forgetting on 6 VidQA datasets spanning various domains.

Continual Learning Image Classification +2

SELMA: Learning and Merging Skill-Specific Text-to-Image Experts with Auto-Generated Data

no code implementations11 Mar 2024 Jialu Li, Jaemin Cho, Yi-Lin Sung, Jaehong Yoon, Mohit Bansal

In this paper, we introduce SELMA: Skill-Specific Expert Learning and Merging with Auto-Generated Data, a novel paradigm to improve the faithfulness of T2I models by fine-tuning models on automatically generated, multi-skill image-text datasets, with skill-specific expert learning and merging.

In-Context Learning

Contrastive Region Guidance: Improving Grounding in Vision-Language Models without Training

no code implementations4 Mar 2024 David Wan, Jaemin Cho, Elias Stengel-Eskin, Mohit Bansal

Highlighting particularly relevant regions of an image can improve the performance of vision-language models (VLMs) on various vision-language (VL) tasks by guiding the model to attend more closely to these regions of interest.

Math Phrase Grounding +3

Evaluating Very Long-Term Conversational Memory of LLM Agents

1 code implementation27 Feb 2024 Adyasha Maharana, Dong-Ho Lee, Sergey Tulyakov, Mohit Bansal, Francesco Barbieri, Yuwei Fang

Using this pipeline, we collect LoCoMo, a dataset of very long-term conversations, each encompassing 300 turns and 9K tokens on avg., over up to 35 sessions.

Avg Multi-modal Dialogue Generation +3

Soft Self-Consistency Improves Language Model Agents

1 code implementation20 Feb 2024 Han Wang, Archiki Prasad, Elias Stengel-Eskin, Mohit Bansal

Current "sample and select" methods such as self-consistency (SC) rely on majority voting to score answers.

Language Modeling Language Modelling +2

Inducing Systematicity in Transformers by Attending to Structurally Quantized Embeddings

1 code implementation9 Feb 2024 Yichen Jiang, Xiang Zhou, Mohit Bansal

Transformers generalize to novel compositions of structures and entities after being trained on a complex dataset, but easily overfit on datasets of insufficient complexity.

Machine Translation Quantization +2

CREMA: Generalizable and Efficient Video-Language Reasoning via Multimodal Modular Fusion

1 code implementation8 Feb 2024 Shoubin Yu, Jaehong Yoon, Mohit Bansal

Despite impressive advancements in recent multimodal reasoning approaches, they are still limited in flexibility and efficiency, as these models typically process only a few fixed modality inputs and require updates to numerous parameters.

Computational Efficiency Multimodal Reasoning +4

VLN-Video: Utilizing Driving Videos for Outdoor Vision-and-Language Navigation

no code implementations5 Feb 2024 Jialu Li, Aishwarya Padmakumar, Gaurav Sukhatme, Mohit Bansal

Outdoor Vision-and-Language Navigation (VLN) requires an agent to navigate through realistic 3D outdoor environments based on natural language instructions.

Language Modeling Language Modelling +3

MAGDi: Structured Distillation of Multi-Agent Interaction Graphs Improves Reasoning in Smaller Language Models

1 code implementation2 Feb 2024 Justin Chih-Yao Chen, Swarnadeep Saha, Elias Stengel-Eskin, Mohit Bansal

Experiments on seven widely used commonsense and math reasoning benchmarks show that MAGDi improves the reasoning capabilities of smaller models, outperforming several methods that distill from a single teacher and multiple teachers.

Language Modelling Large Language Model +1

ReGAL: Refactoring Programs to Discover Generalizable Abstractions

1 code implementation29 Jan 2024 Elias Stengel-Eskin, Archiki Prasad, Mohit Bansal

While large language models (LLMs) are increasingly being used for program synthesis, they lack the global view needed to develop useful abstractions; they generally predict programs one at a time, often repeating the same functionality.

Date Understanding Math +2

Mementos: A Comprehensive Benchmark for Multimodal Large Language Model Reasoning over Image Sequences

1 code implementation19 Jan 2024 Xiyao Wang, YuHang Zhou, Xiaoyu Liu, Hongjin Lu, Yuancheng Xu, Feihong He, Jaehong Yoon, Taixi Lu, Gedas Bertasius, Mohit Bansal, Huaxiu Yao, Furong Huang

However, current MLLM benchmarks are predominantly designed to evaluate reasoning based on static information about a single image, and the ability of modern MLLMs to extrapolate from image sequences, which is essential for understanding our ever-changing world, has been less investigated.

Language Modeling Language Modelling +2

The Unreasonable Effectiveness of Easy Training Data for Hard Tasks

1 code implementation12 Jan 2024 Peter Hase, Mohit Bansal, Peter Clark, Sarah Wiegreffe

In this paper, we present the surprising conclusion that current pretrained language models often generalize relatively well from easy to hard data, even performing as well as oracle models finetuned on hard data.

General Knowledge In-Context Learning +1

Rethinking Interactive Image Segmentation with Low Latency High Quality and Diverse Prompts

1 code implementation CVPR 2024 Qin Liu, Jaemin Cho, Mohit Bansal, Marc Niethammer

In light of this we reintroduce this dense design into the generalist models to facilitate the development of generalist models with high segmentation quality.

Image Segmentation Interactive Segmentation +2

A Simple LLM Framework for Long-Range Video Question-Answering

1 code implementation28 Dec 2023 Ce Zhang, Taixi Lu, Md Mohaiminul Islam, Ziyang Wang, Shoubin Yu, Mohit Bansal, Gedas Bertasius

Furthermore, we show that a specialized prompt that asks the LLM first to summarize the noisy short-term visual captions and then answer a given input question leads to a significant LVQA performance boost.

EgoSchema Large Language Model +3

Merging by Matching Models in Task Parameter Subspaces

1 code implementation7 Dec 2023 Derek Tam, Mohit Bansal, Colin Raffel

Model merging aims to cheaply combine individual task-specific models into a single multitask model.

CoDi-2: In-Context, Interleaved, and Interactive Any-to-Any Generation

no code implementations30 Nov 2023 Zineng Tang, ZiYi Yang, Mahmoud Khademi, Yang Liu, Chenguang Zhu, Mohit Bansal

We present CoDi-2, a versatile and interactive Multimodal Large Language Model (MLLM) that can follow complex multimodal interleaved instructions, conduct in-context learning (ICL), reason, chat, edit, etc., in an any-to-any input-output modality paradigm.

Image Generation In-Context Learning +5

Debiasing Multimodal Models via Causal Information Minimization

1 code implementation28 Nov 2023 Vaidehi Patil, Adyasha Maharana, Mohit Bansal

In this paper, we study bias arising from confounders in a causal graph for multimodal data and examine a novel approach that leverages causally-motivated information minimization to learn the confounder representations.

Visual Question Answering (VQA)

ComPEFT: Compression for Communicating Parameter Efficient Updates via Sparsification and Quantization

1 code implementation22 Nov 2023 Prateek Yadav, Leshem Choshen, Colin Raffel, Mohit Bansal

Despite the efficiency of PEFT methods, the size of expert models can make it onerous to retrieve expert models per query over high-latency networks like the Internet or serve multiple experts on a single GPU.

Language Modelling MMLU +2

Multimodal Representation Learning by Alternating Unimodal Adaptation

no code implementations CVPR 2024 Xiaohui Zhang, Jaehong Yoon, Mohit Bansal, Huaxiu Yao

This optimization process is controlled by a gradient modification mechanism to prevent the shared head from losing previously acquired information.

Representation Learning

ADaPT: As-Needed Decomposition and Planning with Language Models

1 code implementation8 Nov 2023 Archiki Prasad, Alexander Koller, Mareike Hartmann, Peter Clark, Ashish Sabharwal, Mohit Bansal, Tushar Khot

Large Language Models (LLMs) are increasingly being used for interactive decision-making tasks requiring planning and adapting to the environment.

Decision Making

Data Factors for Better Compositional Generalization

1 code implementation8 Nov 2023 Xiang Zhou, Yichen Jiang, Mohit Bansal

However, in contrast to this poor performance, state-of-the-art models trained on larger and more general datasets show better generalization ability.

Diagnostic Memorization

Davidsonian Scene Graph: Improving Reliability in Fine-grained Evaluation for Text-to-Image Generation

no code implementations27 Oct 2023 Jaemin Cho, Yushi Hu, Roopal Garg, Peter Anderson, Ranjay Krishna, Jason Baldridge, Mohit Bansal, Jordi Pont-Tuset, Su Wang

With extensive experimentation and human evaluation on a range of model configurations (LLM, VQA, and T2I), we empirically demonstrate that DSG addresses the challenges noted above.

Question Answering Question Generation +3

Branch-Solve-Merge Improves Large Language Model Evaluation and Generation

no code implementations23 Oct 2023 Swarnadeep Saha, Omer Levy, Asli Celikyilmaz, Mohit Bansal, Jason Weston, Xian Li

Large Language Models (LLMs) are frequently used for multi-faceted language generation and evaluation tasks that involve satisfying intricate user constraints or taking into account multiple aspects and criteria.

Language Model Evaluation Language Modeling +3

DiagrammerGPT: Generating Open-Domain, Open-Platform Diagrams via LLM Planning

no code implementations18 Oct 2023 Abhay Zala, Han Lin, Jaemin Cho, Mohit Bansal

In the second stage, we use a diagram generator, DiagramGLIGEN, and a text label rendering module to generate diagrams (with clear text labels) following the diagram plans.

D2 Pruning: Message Passing for Balancing Diversity and Difficulty in Data Pruning

1 code implementation11 Oct 2023 Adyasha Maharana, Prateek Yadav, Mohit Bansal

There are two dominant approaches: (1) geometry-based data selection for maximizing data diversity in the coreset, and (2) functions that assign difficulty scores to samples based on training dynamics.

Diversity

Rephrase, Augment, Reason: Visual Grounding of Questions for Vision-Language Models

1 code implementation9 Oct 2023 Archiki Prasad, Elias Stengel-Eskin, Mohit Bansal

An increasing number of vision-language tasks can be handled with little to no training, i. e., in a zero and few-shot manner, by marrying large language models (LLMs) to vision encoders, resulting in large vision-language models (LVLMs).

Language Modelling Question Answering +2

ECoFLaP: Efficient Coarse-to-Fine Layer-Wise Pruning for Vision-Language Models

no code implementations4 Oct 2023 Yi-Lin Sung, Jaehong Yoon, Mohit Bansal

We first determine the sparsity ratios of different layers or blocks by leveraging the global importance score, which is efficiently computed based on the zeroth-order approximation of the global model gradients.

Model Compression

Merge, Then Compress: Demystify Efficient SMoE with Hints from Its Routing Policy

1 code implementation2 Oct 2023 Pingzhi Li, Zhenyu Zhang, Prateek Yadav, Yi-Lin Sung, Yu Cheng, Mohit Bansal, Tianlong Chen

Sparsely activated Mixture-of-Experts (SMoE) has shown promise to scale up the learning capacity of neural networks, however, they have issues like (a) High Memory Usage, due to duplication of the network layers into multiple copies as experts; and (b) Redundancy in Experts, as common learning-based routing policies suffer from representational collapse.

Mixture-of-Experts

Can Sensitive Information Be Deleted From LLMs? Objectives for Defending Against Extraction Attacks

1 code implementation29 Sep 2023 Vaidehi Patil, Peter Hase, Mohit Bansal

Experimentally, we show that even state-of-the-art model editing methods such as ROME struggle to truly delete factual information from models like GPT-J, as our whitebox and blackbox attacks can recover "deleted" information from an edited model 38% of the time.

Model Editing

VideoDirectorGPT: Consistent Multi-scene Video Generation via LLM-Guided Planning

no code implementations26 Sep 2023 Han Lin, Abhay Zala, Jaemin Cho, Mohit Bansal

Our experiments demonstrate that our proposed VideoDirectorGPT framework substantially improves layout and movement control in both single- and multi-scene video generation and can generate multi-scene videos with consistency, while achieving competitive performance with SOTAs in open-domain single-scene T2V generation.

Image Generation Video Generation

ReConcile: Round-Table Conference Improves Reasoning via Consensus among Diverse LLMs

2 code implementations22 Sep 2023 Justin Chih-Yao Chen, Swarnadeep Saha, Mohit Bansal

In each round, ReConcile initiates discussion between agents via a 'discussion prompt' that consists of (a) grouped answers and explanations generated by each agent in the previous round, (b) their confidence scores, and (c) demonstrations of answer-rectifying human explanations, used for convincing other agents.

Math

Scaling Data Generation in Vision-and-Language Navigation

1 code implementation ICCV 2023 Zun Wang, Jialu Li, Yicong Hong, Yi Wang, Qi Wu, Mohit Bansal, Stephen Gould, Hao Tan, Yu Qiao

Recent research in language-guided visual navigation has demonstrated a significant demand for the diversity of traversable environments and the quantity of supervision for training generalizable agents.

Imitation Learning Vision and Language Navigation +1

On Conditional and Compositional Language Model Differentiable Prompting

no code implementations4 Jul 2023 Jonathan Pilault, Can Liu, Mohit Bansal, Markus Dreyer

Prompts have been shown to be an effective method to adapt a frozen Pretrained Language Model (PLM) to perform well on downstream tasks.

Few-Shot Learning Language Modeling +3

Can Language Models Teach Weaker Agents? Teacher Explanations Improve Students via Personalization

1 code implementation15 Jun 2023 Swarnadeep Saha, Peter Hase, Mohit Bansal

We first show that teacher LLMs can indeed intervene on student reasoning to improve their performance.

TIES-Merging: Resolving Interference When Merging Models

3 code implementations NeurIPS 2023 Prateek Yadav, Derek Tam, Leshem Choshen, Colin Raffel, Mohit Bansal

To address this, we propose our method, TRIM, ELECT SIGN & MERGE (TIES-Merging), which introduces three novel steps when merging models: (1) resetting parameters that only changed a small amount during fine-tuning, (2) resolving sign conflicts, and (3) merging only the parameters that are in alignment with the final agreed-upon sign.

Transfer Learning

Non-Sequential Graph Script Induction via Multimedia Grounding

1 code implementation27 May 2023 Yu Zhou, Sha Li, Manling Li, Xudong Lin, Shih-Fu Chang, Mohit Bansal, Heng Ji

To automate the induction of such graph scripts for given tasks, we propose to take advantage of loosely aligned videos of people performing the tasks.

MixCE: Training Autoregressive Language Models by Mixing Forward and Reverse Cross-Entropies

1 code implementation26 May 2023 Shiyue Zhang, Shijie Wu, Ozan Irsoy, Steven Lu, Mohit Bansal, Mark Dredze, David Rosenberg

Autoregressive language models are trained by minimizing the cross-entropy of the model distribution Q relative to the data distribution P -- that is, minimizing the forward cross-entropy, which is equivalent to maximum likelihood estimation (MLE).

Visual Programming for Text-to-Image Generation and Evaluation

no code implementations24 May 2023 Jaemin Cho, Abhay Zala, Mohit Bansal

First, we introduce VPGen, an interpretable step-by-step T2I generation framework that decomposes T2I generation into three steps: object/count generation, layout generation, and image generation.

Layout Generation Text-to-Image Generation +1

Any-to-Any Generation via Composable Diffusion

2 code implementations NeurIPS 2023 Zineng Tang, ZiYi Yang, Chenguang Zhu, Michael Zeng, Mohit Bansal

We present Composable Diffusion (CoDi), a novel generative model capable of generating any combination of output modalities, such as language, image, video, or audio, from any combination of input modalities.

Ranked #8 on Audio Generation on AudioCaps (FAD metric)

Audio Generation

HistAlign: Improving Context Dependency in Language Generation by Aligning with History

1 code implementation8 May 2023 David Wan, Shiyue Zhang, Mohit Bansal

Cache-LMs, which augment LMs with a memory of recent history, can increase context dependency and have shown remarkable performance in diverse language generation tasks.

Abstractive Text Summarization Text Generation

An Empirical Study of Multimodal Model Merging

1 code implementation28 Apr 2023 Yi-Lin Sung, Linjie Li, Kevin Lin, Zhe Gan, Mohit Bansal, Lijuan Wang

In this paper, we expand on this concept to a multimodal setup by merging transformers trained on different modalities.

model Retrieval +2

ReCEval: Evaluating Reasoning Chains via Correctness and Informativeness

1 code implementation21 Apr 2023 Archiki Prasad, Swarnadeep Saha, Xiang Zhou, Mohit Bansal

Multi-step reasoning ability is fundamental to many natural language tasks, yet it is unclear what constitutes a good reasoning chain and how to evaluate them.

Informativeness Natural Language Inference +1

Diagnostic Benchmark and Iterative Inpainting for Layout-Guided Image Generation

2 code implementations13 Apr 2023 Jaemin Cho, Linjie Li, Zhengyuan Yang, Zhe Gan, Lijuan Wang, Mohit Bansal

In this paper, we propose LayoutBench, a diagnostic benchmark for layout-guided image generation that examines four categories of spatial control skills: number, position, size, and shape.

Diagnostic Layout-to-Image Generation

Improving Vision-and-Language Navigation by Generating Future-View Image Semantics

no code implementations CVPR 2023 Jialu Li, Mohit Bansal

We then fine-tune the agent on the VLN task with an auxiliary loss that minimizes the difference between the view semantics generated by the agent and the ground truth view semantics of the next step.

Image Generation Navigate +3

Hierarchical Video-Moment Retrieval and Step-Captioning

1 code implementation CVPR 2023 Abhay Zala, Jaemin Cho, Satwik Kottur, Xilun Chen, Barlas Oğuz, Yasher Mehdad, Mohit Bansal

Our hierarchical benchmark consists of video retrieval, moment retrieval, and two novel moment segmentation and step captioning tasks.

Information Retrieval Moment Retrieval +4

Exposing and Addressing Cross-Task Inconsistency in Unified Vision-Language Models

1 code implementation28 Mar 2023 Adyasha Maharana, Amita Kamath, Christopher Clark, Mohit Bansal, Aniruddha Kembhavi

As general purpose vision models get increasingly effective at a wide set of tasks, it is imperative that they be consistent across the tasks they support.

Faithfulness-Aware Decoding Strategies for Abstractive Summarization

1 code implementation6 Mar 2023 David Wan, Mengwen Liu, Kathleen McKeown, Markus Dreyer, Mohit Bansal

We present a systematic study of the effect of generation techniques such as beam search and nucleus sampling on faithfulness in abstractive summarization.

Abstractive Text Summarization

Vision Transformers are Parameter-Efficient Audio-Visual Learners

1 code implementation CVPR 2023 Yan-Bo Lin, Yi-Lin Sung, Jie Lei, Mohit Bansal, Gedas Bertasius

To do so, we propose a latent audio-visual hybrid (LAVISH) adapter that adapts pretrained ViTs to audio-visual tasks by injecting a small number of trainable parameters into every layer of a frozen ViT.

Audio-visual Question Answering

VindLU: A Recipe for Effective Video-and-Language Pretraining

1 code implementation CVPR 2023 Feng Cheng, Xizi Wang, Jie Lei, David Crandall, Mohit Bansal, Gedas Bertasius

Furthermore, our model also obtains state-of-the-art video question-answering results on ActivityNet-QA, MSRVTT-QA, MSRVTT-MC and TVQA.

Ranked #2 on Video Retrieval on Condensed Movies (using extra training data)

Question Answering Retrieval +3

Unifying Vision, Text, and Layout for Universal Document Processing

4 code implementations CVPR 2023 Zineng Tang, ZiYi Yang, Guoxin Wang, Yuwei Fang, Yang Liu, Chenguang Zhu, Michael Zeng, Cha Zhang, Mohit Bansal

UDOP leverages the spatial correlation between textual content and document image to model image, text, and layout modalities with one uniform representation.

Ranked #5 on Visual Question Answering (VQA) on InfographicVQA (using extra training data)

document understanding Image Reconstruction +1

Mutual Exclusivity Training and Primitive Augmentation to Induce Compositionality

1 code implementation28 Nov 2022 Yichen Jiang, Xiang Zhou, Mohit Bansal

Recent datasets expose the lack of the systematic generalization ability in standard sequence-to-sequence models.

Data Augmentation Inductive Bias +1

Perceiver-VL: Efficient Vision-and-Language Modeling with Iterative Latent Attention

1 code implementation21 Nov 2022 Zineng Tang, Jaemin Cho, Jie Lei, Mohit Bansal

We present Perceiver-VL, a vision-and-language framework that efficiently handles high-dimensional multimodal inputs such as long videos and text.

Cross-Modal Retrieval Language Modeling +2

Evaluating the Factual Consistency of Large Language Models Through News Summarization

1 code implementation15 Nov 2022 Derek Tam, Anisha Mascarenhas, Shiyue Zhang, Sarah Kwan, Mohit Bansal, Colin Raffel

To generate summaries that are factually inconsistent, we generate summaries from a suite of summarization models that we have manually annotated as factually inconsistent.

News Summarization

Are Hard Examples also Harder to Explain? A Study with Human and Model-Generated Explanations

1 code implementation14 Nov 2022 Swarnadeep Saha, Peter Hase, Nazneen Rajani, Mohit Bansal

We observe that (1) GPT-3 explanations are as grammatical as human explanations regardless of the hardness of the test samples, (2) for easy examples, GPT-3 generates highly supportive explanations but human explanations are more generalizable, and (3) for hard examples, human explanations are significantly better than GPT-3 explanations both in terms of label-supportiveness and generalizability judgements.

Winogrande

Evaluating and Improving Factuality in Multimodal Abstractive Summarization

1 code implementation4 Nov 2022 David Wan, Mohit Bansal

Current metrics for evaluating factuality for abstractive document summarization have achieved high correlations with human judgment, but they do not account for the vision modality and thus are not adequate for vision-and-language summarization.

Abstractive Text Summarization Document Summarization

Exclusive Supermask Subnetwork Training for Continual Learning

1 code implementation18 Oct 2022 Prateek Yadav, Mohit Bansal

Although there is no forgetting, the performance of SupSup is sub-optimal because fixed weights restrict its representational power.

Continual Learning Text Classification +1

TVLT: Textless Vision-Language Transformer

3 code implementations28 Sep 2022 Zineng Tang, Jaemin Cho, Yixin Nie, Mohit Bansal

In this work, we present the Textless Vision-Language Transformer (TVLT), where homogeneous transformer blocks take raw visual and audio inputs for vision-and-language representation learning with minimal modality-specific design, and do not use text-specific modules such as tokenization or automatic speech recognition (ASR).

Automatic Speech Recognition (ASR) Image Retrieval +6

Summarization Programs: Interpretable Abstractive Summarization with Neural Modular Trees

1 code implementation21 Sep 2022 Swarnadeep Saha, Shiyue Zhang, Peter Hase, Mohit Bansal

We demonstrate that SP-Search effectively represents the generative process behind human summaries using modules that are typically faithful to their intended behavior.

Abstractive Text Summarization Sentence +1

StoryDALL-E: Adapting Pretrained Text-to-Image Transformers for Story Continuation

1 code implementation13 Sep 2022 Adyasha Maharana, Darryl Hannan, Mohit Bansal

Hence, we first propose the task of story continuation, where the generated visual story is conditioned on a source image, allowing for better generalization to narratives with new characters.

Image Generation Story Continuation +2

Extractive is not Faithful: An Investigation of Broad Unfaithfulness Problems in Extractive Summarization

1 code implementation8 Sep 2022 Shiyue Zhang, David Wan, Mohit Bansal

Though extractive summarization is less prone to the common unfaithfulness issues of abstractive summaries, does that mean extractive is equal to faithful?

Abstractive Text Summarization Extractive Summarization

WinoGAViL: Gamified Association Benchmark to Challenge Vision-and-Language Models

1 code implementation25 Jul 2022 Yonatan Bitton, Nitzan Bitton Guetta, Ron Yosef, Yuval Elovici, Mohit Bansal, Gabriel Stanovsky, Roy Schwartz

While vision-and-language models perform well on tasks such as visual question answering, they struggle when it comes to basic human commonsense reasoning skills.

Common Sense Reasoning General Knowledge +4

CoSIm: Commonsense Reasoning for Counterfactual Scene Imagination

1 code implementation NAACL 2022 Hyounghun Kim, Abhay Zala, Mohit Bansal

Next, a counterfactual imagined scene change (in textual form) is applied, and the model has to predict the new response to the initial question based on this scene change.

counterfactual

SETSum: Summarization and Visualization of Student Evaluations of Teaching

1 code implementation NAACL (ACL) 2022 Yinuo Hu, Shiyue Zhang, Viji Sathy, A. T. Panter, Mohit Bansal

Ten university professors from diverse departments serve as evaluators of the system and all agree that SETSum helps them interpret SET results more efficiently; and 6 out of 10 instructors prefer our system over the standard static PDF report (while the remaining 4 would like to have both).

Aspect Extraction Sentiment Analysis

CLEAR: Improving Vision-Language Navigation with Cross-Lingual, Environment-Agnostic Representations

1 code implementation Findings (NAACL) 2022 Jialu Li, Hao Tan, Mohit Bansal

Empirically, on the Room-Across-Room dataset, we show that our multilingual agent gets large improvements in all metrics over the strong baseline model when generalizing to unseen environments with the cross-lingual language representation and the environment-agnostic visual representation.

Navigate Representation Learning +2

Masked Part-Of-Speech Model: Does Modeling Long Context Help Unsupervised POS-tagging?

1 code implementation NAACL 2022 Xiang Zhou, Shiyue Zhang, Mohit Bansal

MPoSM can model arbitrary tag dependency and perform POS induction through the objective of masked POS reconstruction.

POS POS Tagging +1

VisFIS: Visual Feature Importance Supervision with Right-for-the-Right-Reason Objectives

1 code implementation22 Jun 2022 Zhuofan Ying, Peter Hase, Mohit Bansal

In this paper, we show that model FI supervision can meaningfully improve VQA model accuracy as well as performance on several Right-for-the-Right-Reason (RRR) metrics by optimizing for four key model objectives: (1) accurate predictions given limited but sufficient information (Sufficiency); (2) max-entropy predictions given no important information (Uncertainty); (3) invariance of predictions to changes in unimportant features (Invariance); and (4) alignment between model FI explanations and human FI explanations (Plausibility).

Feature Importance Question Answering +2

Enhanced Knowledge Selection for Grounded Dialogues via Document Semantic Graphs

no code implementations15 Jun 2022 Sha Li, Mahdi Namazifar, Di Jin, Mohit Bansal, Heng Ji, Yang Liu, Dilek Hakkani-Tur

Providing conversation models with background knowledge has been shown to make open-domain dialogues more informative and engaging.

Multi-Task Learning Response Generation +1

LST: Ladder Side-Tuning for Parameter and Memory Efficient Transfer Learning

2 code implementations13 Jun 2022 Yi-Lin Sung, Jaemin Cho, Mohit Bansal

LST saves 69% of the memory costs to fine-tune the whole network, while other methods only save 26% of that in similar parameter usages (hence, 2. 7x more memory savings).

Transfer Learning Visual Question Answering (VQA)

Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language models

5 code implementations9 Jun 2022 Aarohi Srivastava, Abhinav Rastogi, Abhishek Rao, Abu Awal Md Shoeb, Abubakar Abid, Adam Fisch, Adam R. Brown, Adam Santoro, Aditya Gupta, Adrià Garriga-Alonso, Agnieszka Kluska, Aitor Lewkowycz, Akshat Agarwal, Alethea Power, Alex Ray, Alex Warstadt, Alexander W. Kocurek, Ali Safaya, Ali Tazarv, Alice Xiang, Alicia Parrish, Allen Nie, Aman Hussain, Amanda Askell, Amanda Dsouza, Ambrose Slone, Ameet Rahane, Anantharaman S. Iyer, Anders Andreassen, Andrea Madotto, Andrea Santilli, Andreas Stuhlmüller, Andrew Dai, Andrew La, Andrew Lampinen, Andy Zou, Angela Jiang, Angelica Chen, Anh Vuong, Animesh Gupta, Anna Gottardi, Antonio Norelli, Anu Venkatesh, Arash Gholamidavoodi, Arfa Tabassum, Arul Menezes, Arun Kirubarajan, Asher Mullokandov, Ashish Sabharwal, Austin Herrick, Avia Efrat, Aykut Erdem, Ayla Karakaş, B. Ryan Roberts, Bao Sheng Loe, Barret Zoph, Bartłomiej Bojanowski, Batuhan Özyurt, Behnam Hedayatnia, Behnam Neyshabur, Benjamin Inden, Benno Stein, Berk Ekmekci, Bill Yuchen Lin, Blake Howald, Bryan Orinion, Cameron Diao, Cameron Dour, Catherine Stinson, Cedrick Argueta, César Ferri Ramírez, Chandan Singh, Charles Rathkopf, Chenlin Meng, Chitta Baral, Chiyu Wu, Chris Callison-Burch, Chris Waites, Christian Voigt, Christopher D. Manning, Christopher Potts, Cindy Ramirez, Clara E. Rivera, Clemencia Siro, Colin Raffel, Courtney Ashcraft, Cristina Garbacea, Damien Sileo, Dan Garrette, Dan Hendrycks, Dan Kilman, Dan Roth, Daniel Freeman, Daniel Khashabi, Daniel Levy, Daniel Moseguí González, Danielle Perszyk, Danny Hernandez, Danqi Chen, Daphne Ippolito, Dar Gilboa, David Dohan, David Drakard, David Jurgens, Debajyoti Datta, Deep Ganguli, Denis Emelin, Denis Kleyko, Deniz Yuret, Derek Chen, Derek Tam, Dieuwke Hupkes, Diganta Misra, Dilyar Buzan, Dimitri Coelho Mollo, Diyi Yang, Dong-Ho Lee, Dylan Schrader, Ekaterina Shutova, Ekin Dogus Cubuk, Elad Segal, Eleanor Hagerman, Elizabeth Barnes, Elizabeth Donoway, Ellie Pavlick, Emanuele Rodola, Emma Lam, Eric Chu, Eric Tang, Erkut Erdem, Ernie Chang, Ethan A. Chi, Ethan Dyer, Ethan Jerzak, Ethan Kim, Eunice Engefu Manyasi, Evgenii Zheltonozhskii, Fanyue Xia, Fatemeh Siar, Fernando Martínez-Plumed, Francesca Happé, Francois Chollet, Frieda Rong, Gaurav Mishra, Genta Indra Winata, Gerard de Melo, Germán Kruszewski, Giambattista Parascandolo, Giorgio Mariani, Gloria Wang, Gonzalo Jaimovitch-López, Gregor Betz, Guy Gur-Ari, Hana Galijasevic, Hannah Kim, Hannah Rashkin, Hannaneh Hajishirzi, Harsh Mehta, Hayden Bogar, Henry Shevlin, Hinrich Schütze, Hiromu Yakura, Hongming Zhang, Hugh Mee Wong, Ian Ng, Isaac Noble, Jaap Jumelet, Jack Geissinger, Jackson Kernion, Jacob Hilton, Jaehoon Lee, Jaime Fernández Fisac, James B. Simon, James Koppel, James Zheng, James Zou, Jan Kocoń, Jana Thompson, Janelle Wingfield, Jared Kaplan, Jarema Radom, Jascha Sohl-Dickstein, Jason Phang, Jason Wei, Jason Yosinski, Jekaterina Novikova, Jelle Bosscher, Jennifer Marsh, Jeremy Kim, Jeroen Taal, Jesse Engel, Jesujoba Alabi, Jiacheng Xu, Jiaming Song, Jillian Tang, Joan Waweru, John Burden, John Miller, John U. Balis, Jonathan Batchelder, Jonathan Berant, Jörg Frohberg, Jos Rozen, Jose Hernandez-Orallo, Joseph Boudeman, Joseph Guerr, Joseph Jones, Joshua B. Tenenbaum, Joshua S. Rule, Joyce Chua, Kamil Kanclerz, Karen Livescu, Karl Krauth, Karthik Gopalakrishnan, Katerina Ignatyeva, Katja Markert, Kaustubh D. Dhole, Kevin Gimpel, Kevin Omondi, Kory Mathewson, Kristen Chiafullo, Ksenia Shkaruta, Kumar Shridhar, Kyle McDonell, Kyle Richardson, Laria Reynolds, Leo Gao, Li Zhang, Liam Dugan, Lianhui Qin, Lidia Contreras-Ochando, Louis-Philippe Morency, Luca Moschella, Lucas Lam, Lucy Noble, Ludwig Schmidt, Luheng He, Luis Oliveros Colón, Luke Metz, Lütfi Kerem Şenel, Maarten Bosma, Maarten Sap, Maartje ter Hoeve, Maheen Farooqi, Manaal Faruqui, Mantas Mazeika, Marco Baturan, Marco Marelli, Marco Maru, Maria Jose Ramírez Quintana, Marie Tolkiehn, Mario Giulianelli, Martha Lewis, Martin Potthast, Matthew L. Leavitt, Matthias Hagen, Mátyás Schubert, Medina Orduna Baitemirova, Melody Arnaud, Melvin McElrath, Michael A. Yee, Michael Cohen, Michael Gu, Michael Ivanitskiy, Michael Starritt, Michael Strube, Michał Swędrowski, Michele Bevilacqua, Michihiro Yasunaga, Mihir Kale, Mike Cain, Mimee Xu, Mirac Suzgun, Mitch Walker, Mo Tiwari, Mohit Bansal, Moin Aminnaseri, Mor Geva, Mozhdeh Gheini, Mukund Varma T, Nanyun Peng, Nathan A. Chi, Nayeon Lee, Neta Gur-Ari Krakover, Nicholas Cameron, Nicholas Roberts, Nick Doiron, Nicole Martinez, Nikita Nangia, Niklas Deckers, Niklas Muennighoff, Nitish Shirish Keskar, Niveditha S. Iyer, Noah Constant, Noah Fiedel, Nuan Wen, Oliver Zhang, Omar Agha, Omar Elbaghdadi, Omer Levy, Owain Evans, Pablo Antonio Moreno Casares, Parth Doshi, Pascale Fung, Paul Pu Liang, Paul Vicol, Pegah Alipoormolabashi, Peiyuan Liao, Percy Liang, Peter Chang, Peter Eckersley, Phu Mon Htut, Pinyu Hwang, Piotr Miłkowski, Piyush Patil, Pouya Pezeshkpour, Priti Oli, Qiaozhu Mei, Qing Lyu, Qinlang Chen, Rabin Banjade, Rachel Etta Rudolph, Raefer Gabriel, Rahel Habacker, Ramon Risco, Raphaël Millière, Rhythm Garg, Richard Barnes, Rif A. Saurous, Riku Arakawa, Robbe Raymaekers, Robert Frank, Rohan Sikand, Roman Novak, Roman Sitelew, Ronan LeBras, Rosanne Liu, Rowan Jacobs, Rui Zhang, Ruslan Salakhutdinov, Ryan Chi, Ryan Lee, Ryan Stovall, Ryan Teehan, Rylan Yang, Sahib Singh, Saif M. Mohammad, Sajant Anand, Sam Dillavou, Sam Shleifer, Sam Wiseman, Samuel Gruetter, Samuel R. Bowman, Samuel S. Schoenholz, Sanghyun Han, Sanjeev Kwatra, Sarah A. Rous, Sarik Ghazarian, Sayan Ghosh, Sean Casey, Sebastian Bischoff, Sebastian Gehrmann, Sebastian Schuster, Sepideh Sadeghi, Shadi Hamdan, Sharon Zhou, Shashank Srivastava, Sherry Shi, Shikhar Singh, Shima Asaadi, Shixiang Shane Gu, Shubh Pachchigar, Shubham Toshniwal, Shyam Upadhyay, Shyamolima, Debnath, Siamak Shakeri, Simon Thormeyer, Simone Melzi, Siva Reddy, Sneha Priscilla Makini, Soo-Hwan Lee, Spencer Torene, Sriharsha Hatwar, Stanislas Dehaene, Stefan Divic, Stefano Ermon, Stella Biderman, Stephanie Lin, Stephen Prasad, Steven T. Piantadosi, Stuart M. Shieber, Summer Misherghi, Svetlana Kiritchenko, Swaroop Mishra, Tal Linzen, Tal Schuster, Tao Li, Tao Yu, Tariq Ali, Tatsu Hashimoto, Te-Lin Wu, Théo Desbordes, Theodore Rothschild, Thomas Phan, Tianle Wang, Tiberius Nkinyili, Timo Schick, Timofei Kornev, Titus Tunduny, Tobias Gerstenberg, Trenton Chang, Trishala Neeraj, Tushar Khot, Tyler Shultz, Uri Shaham, Vedant Misra, Vera Demberg, Victoria Nyamai, Vikas Raunak, Vinay Ramasesh, Vinay Uday Prabhu, Vishakh Padmakumar, Vivek Srikumar, William Fedus, William Saunders, William Zhang, Wout Vossen, Xiang Ren, Xiaoyu Tong, Xinran Zhao, Xinyi Wu, Xudong Shen, Yadollah Yaghoobzadeh, Yair Lakretz, Yangqiu Song, Yasaman Bahri, Yejin Choi, Yichi Yang, Yiding Hao, Yifu Chen, Yonatan Belinkov, Yu Hou, Yufang Hou, Yuntao Bai, Zachary Seid, Zhuoye Zhao, Zijian Wang, Zijie J. Wang, ZiRui Wang, Ziyi Wu

BIG-bench focuses on tasks that are believed to be beyond the capabilities of current language models.

Common Sense Reasoning Math +1

Revealing Single Frame Bias for Video-and-Language Learning

2 code implementations7 Jun 2022 Jie Lei, Tamara L. Berg, Mohit Bansal

Training an effective video-and-language model intuitively requires multiple frames as model inputs.

Ranked #5 on Video Retrieval on SSv2-template retrieval (using extra training data)

Fine-grained Action Recognition Language Modeling +7

Fine-grained Image Captioning with CLIP Reward

1 code implementation Findings (NAACL) 2022 Jaemin Cho, Seunghyun Yoon, Ajinkya Kale, Franck Dernoncourt, Trung Bui, Mohit Bansal

Toward more descriptive and distinctive caption generation, we propose using CLIP, a multimodal encoder trained on huge image-text pairs from web, to calculate multimodal similarity and use it as a reward function.

Caption Generation Descriptive +5

Language Models with Image Descriptors are Strong Few-Shot Video-Language Learners

1 code implementation22 May 2022 Zhenhailong Wang, Manling Li, Ruochen Xu, Luowei Zhou, Jie Lei, Xudong Lin, Shuohang Wang, ZiYi Yang, Chenguang Zhu, Derek Hoiem, Shih-Fu Chang, Mohit Bansal, Heng Ji

The goal of this work is to build flexible video-language models that can generalize to various video-to-text tasks from few examples, such as domain-specific captioning, question answering, and future event prediction.

Attribute Automatic Speech Recognition +6

On the Limits of Evaluating Embodied Agent Model Generalization Using Validation Sets

no code implementations insights (ACL) 2022 Hyounghun Kim, Aishwarya Padmakumar, Di Jin, Mohit Bansal, Dilek Hakkani-Tur

Natural language guided embodied task completion is a challenging problem since it requires understanding natural language instructions, aligning them with egocentric visual observations, and choosing appropriate actions to execute in the environment to produce desired changes.

FactPEGASUS: Factuality-Aware Pre-training and Fine-tuning for Abstractive Summarization

1 code implementation NAACL 2022 David Wan, Mohit Bansal

We present FactPEGASUS, an abstractive summarization model that addresses the problem of factuality during pre-training and fine-tuning: (1) We augment the sentence selection strategy of PEGASUS's (Zhang et al., 2020) pre-training objective to create pseudo-summaries that are both important and factual; (2) We introduce three complementary components for fine-tuning.

Abstractive Text Summarization Contrastive Learning +1

Few-Shot Parameter-Efficient Fine-Tuning is Better and Cheaper than In-Context Learning

2 code implementations11 May 2022 Haokun Liu, Derek Tam, Mohammed Muqeeth, Jay Mohta, Tenghao Huang, Mohit Bansal, Colin Raffel

ICL incurs substantial computational, memory, and storage costs because it involves processing all of the training examples every time a prediction is made.

Few-Shot Text Classification In-Context Learning +1

How can NLP Help Revitalize Endangered Languages? A Case Study and Roadmap for the Cherokee Language

1 code implementation ACL 2022 Shiyue Zhang, Ben Frey, Mohit Bansal

We hope that our work serves not only to inform the NLP community about Cherokee, but also to provide inspiration for future work on endangered languages in general.

FactGraph: Evaluating Factuality in Summarization with Semantic Graph Representations

3 code implementations NAACL 2022 Leonardo F. R. Ribeiro, Mengwen Liu, Iryna Gurevych, Markus Dreyer, Mohit Bansal

Despite recent improvements in abstractive summarization, most current approaches generate summaries that are not factually consistent with the source document, severely restricting their trust and usage in real-world applications.

Abstractive Text Summarization ARC

Explanation Graph Generation via Pre-trained Language Models: An Empirical Study with Contrastive Learning

1 code implementation ACL 2022 Swarnadeep Saha, Prateek Yadav, Mohit Bansal

In this work, we study pre-trained language models that generate explanation graphs in an end-to-end manner and analyze their ability to learn the structural constraints and semantics of such graphs.

Contrastive Learning Graph Generation +1

EnvEdit: Environment Editing for Vision-and-Language Navigation

1 code implementation CVPR 2022 Jialu Li, Hao Tan, Mohit Bansal

Training on these edit-augmented environments prevents the agent from overfitting to existing environments and helps generalize better to new, unseen environments.

Ranked #2 on Vision and Language Navigation on RxR (using extra training data)

Data Augmentation Diversity +2

GrIPS: Gradient-free, Edit-based Instruction Search for Prompting Large Language Models

2 code implementations14 Mar 2022 Archiki Prasad, Peter Hase, Xiang Zhou, Mohit Bansal

Providing natural language instructions in prompts is a useful new paradigm for improving task performance of large language models in a zero-shot setting.

CAISE: Conversational Agent for Image Search and Editing

1 code implementation24 Feb 2022 Hyounghun Kim, Doo Soon Kim, Seunghyun Yoon, Franck Dernoncourt, Trung Bui, Mohit Bansal

To our knowledge, this is the first dataset that provides conversational image search and editing annotations, where the agent holds a grounded conversation with users and helps them to search and edit images according to their requests.

Image Retrieval

DALL-Eval: Probing the Reasoning Skills and Social Biases of Text-to-Image Generation Models

2 code implementations ICCV 2023 Jaemin Cho, Abhay Zala, Mohit Bansal

In this work, we investigate the visual reasoning capabilities and social biases of different text-to-image models, covering both multimodal transformer language models and diffusion models.

Diagnostic Image Captioning +9

MuMuQA: Multimedia Multi-Hop News Question Answering via Cross-Media Knowledge Extraction and Grounding

2 code implementations20 Dec 2021 Revanth Gangi Reddy, Xilin Rui, Manling Li, Xudong Lin, Haoyang Wen, Jaemin Cho, Lifu Huang, Mohit Bansal, Avirup Sil, Shih-Fu Chang, Alexander Schwing, Heng Ji

Specifically, the task involves multi-hop questions that require reasoning over image-caption pairs to identify the grounded visual object being referred to and then predicting a span from the news body text to answer the question.

Answer Generation Data Augmentation +2

Proposition-Level Clustering for Multi-Document Summarization

2 code implementations NAACL 2022 Ori Ernst, Avi Caciularu, Ori Shapira, Ramakanth Pasunuru, Mohit Bansal, Jacob Goldberger, Ido Dagan

Text clustering methods were traditionally incorporated into multi-document summarization (MDS) as a means for coping with considerable information repetition.

Clustering Document Summarization +3

Learning and Analyzing Generation Order for Undirected Sequence Models

1 code implementation Findings (EMNLP) 2021 Yichen Jiang, Mohit Bansal

On examples with a maximum source and target length of 30 from De-En, WMT'16 English-Romanian, and WMT'21 English-Chinese translation tasks, our learned order outperforms all heuristic generation orders on four out of six tasks.

de-en Machine Translation +1

Analyzing the Limits of Self-Supervision in Handling Bias in Language

no code implementations16 Dec 2021 Lisa Bauer, Karthik Gopalakrishnan, Spandana Gella, Yang Liu, Mohit Bansal, Dilek Hakkani-Tur

We define three broad classes of task descriptions for these tasks: statement, question, and completion, with numerous lexical variants within each class.

VL-Adapter: Parameter-Efficient Transfer Learning for Vision-and-Language Tasks

1 code implementation CVPR 2022 Yi-Lin Sung, Jaemin Cho, Mohit Bansal

Our results demonstrate that training the adapter with the weight-sharing technique (4. 18% of total parameters for image-text tasks and 3. 39% for video-text tasks) can match the performance of fine-tuning the entire model.

Image Captioning Transfer Learning

MLP Architectures for Vision-and-Language Modeling: An Empirical Study

1 code implementation8 Dec 2021 Yixin Nie, Linjie Li, Zhe Gan, Shuohang Wang, Chenguang Zhu, Michael Zeng, Zicheng Liu, Mohit Bansal, Lijuan Wang

Based on this, we ask an even bolder question: can we have an all-MLP architecture for VL modeling, where both VL fusion and the vision encoder are replaced with MLPs?

Language Modeling Language Modelling +1

Detecting Moments and Highlights in Videos via Natural Language Queries

1 code implementation NeurIPS 2021 Jie Lei, Tamara Berg, Mohit Bansal

Each video in the dataset is annotated with: (1) a human-written free-form NL query, (2) relevant moments in the video w. r. t.

Decoder Moment Retrieval +3

Do Language Models Have Beliefs? Methods for Detecting, Updating, and Visualizing Model Beliefs

1 code implementation26 Nov 2021 Peter Hase, Mona Diab, Asli Celikyilmaz, Xian Li, Zornitsa Kozareva, Veselin Stoyanov, Mohit Bansal, Srinivasan Iyer

In this paper, we discuss approaches to detecting when models have beliefs about the world, and we improve on methods for updating model beliefs to be more truthful, with a focus on methods based on learned optimizers or hypernetworks.

Low-Cost Algorithmic Recourse for Users With Uncertain Cost Functions

1 code implementation1 Nov 2021 Prateek Yadav, Peter Hase, Mohit Bansal

Current approaches try to optimize for the cost incurred by users when adopting a recourse, but they assume that all users share the same cost function.

Fairness

Integrating Visuospatial, Linguistic and Commonsense Structure into Story Visualization

1 code implementation21 Oct 2021 Adyasha Maharana, Mohit Bansal

Prior work in this domain has shown that there is ample room for improvement in the generated image sequence in terms of visual quality, consistency and relevance.

Dense Captioning Image Generation +1

Inducing Transformer's Compositional Generalization Ability via Auxiliary Sequence Prediction Tasks

1 code implementation30 Sep 2021 Yichen Jiang, Mohit Bansal

Motivated by the failure of a Transformer model on the SCAN compositionality challenge (Lake and Baroni, 2018), which requires parsing a command into actions, we propose two auxiliary sequence prediction tasks that track the progress of function and argument semantics, as additional training supervision.

Finding a Balanced Degree of Automation for Summary Evaluation

1 code implementation EMNLP 2021 Shiyue Zhang, Mohit Bansal

In this work, we propose flexible semiautomatic to automatic summary evaluation metrics, following the Pyramid human evaluation method.

Natural Language Inference Semantic Role Labeling +1

Continuous Language Generative Flow

1 code implementation ACL 2021 Zineng Tang, Shiyue Zhang, Hyounghun Kim, Mohit Bansal

Recent years have witnessed various types of generative models for natural language generation (NLG), especially RNNs or transformer based sequence-to-sequence models, as well as variational autoencoder (VAE) and generative adversarial network (GAN) based models.

Data Augmentation Density Estimation +9

MTVR: Multilingual Moment Retrieval in Videos

1 code implementation ACL 2021 Jie Lei, Tamara L. Berg, Mohit Bansal

We introduce mTVR, a large-scale multilingual video moment retrieval dataset, containing 218K English and Chinese queries from 21. 8K TV show video clips.

Moment Retrieval Retrieval

EmailSum: Abstractive Email Thread Summarization

1 code implementation ACL 2021 Shiyue Zhang, Asli Celikyilmaz, Jianfeng Gao, Mohit Bansal

Furthermore, we find that widely used automatic evaluation metrics (ROUGE, BERTScore) are weakly correlated with human judgments on this email thread summarization task.

Abstractive Text Summarization Email Thread Summarization

ChrEnTranslate: Cherokee-English Machine Translation Demo with Quality Estimation and Corrective Feedback

2 code implementations ACL 2021 Shiyue Zhang, Benjamin Frey, Mohit Bansal

The quantitative evaluation demonstrates that our backbone translation models achieve state-of-the-art translation performance and our quality estimation well correlates with both BLEU and human judgment.

Machine Translation NMT +3

QVHighlights: Detecting Moments and Highlights in Videos via Natural Language Queries

4 code implementations20 Jul 2021 Jie Lei, Tamara L. Berg, Mohit Bansal

Each video in the dataset is annotated with: (1) a human-written free-form NL query, (2) relevant moments in the video w. r. t.

Highlight Detection Moment Retrieval +2

How Much Can CLIP Benefit Vision-and-Language Tasks?

4 code implementations13 Jul 2021 Sheng Shen, Liunian Harold Li, Hao Tan, Mohit Bansal, Anna Rohrbach, Kai-Wei Chang, Zhewei Yao, Kurt Keutzer

Most existing Vision-and-Language (V&L) models rely on pre-trained visual encoders, using a relatively small set of manually-annotated data (as compared to web-crawled data), to perceive the visual world.

Ranked #4 on Vision and Language Navigation on RxR (using extra training data)

Question Answering Vision and Language Navigation +2

VidLanKD: Improving Language Understanding via Video-Distilled Knowledge Transfer

1 code implementation NeurIPS 2021 Zineng Tang, Jaemin Cho, Hao Tan, Mohit Bansal

We train a multi-modal teacher model on a video-text dataset, and then transfer its knowledge to a student language model with a text dataset.

Image Retrieval Knowledge Distillation +7

VIMPAC: Video Pre-Training via Masked Token Prediction and Contrastive Learning

1 code implementation21 Jun 2021 Hao Tan, Jie Lei, Thomas Wolf, Mohit Bansal

Unlike language, where the text tokens are more independent, neighboring video tokens typically have strong correlations (e. g., consecutive video frames usually look very similar), and hence uniformly masking individual tokens will make the task too trivial to learn useful representations.

Action Classification Action Recognition +2

An Empirical Survey of Data Augmentation for Limited Data Learning in NLP

no code implementations14 Jun 2021 Jiaao Chen, Derek Tam, Colin Raffel, Mohit Bansal, Diyi Yang

NLP has achieved great progress in the past decade through the use of neural models and large labeled datasets.

Data Augmentation News Classification +1

multiPRover: Generating Multiple Proofs for Improved Interpretability in Rule Reasoning

1 code implementation NAACL 2021 Swarnadeep Saha, Prateek Yadav, Mohit Bansal

In order to jointly learn from all proof graphs and exploit the correlations between multiple proofs for a question, we pose this task as a set generation problem over structured output spaces where each proof is represented as a directed graph.

Multi-Label Classification MUlTI-LABEL-ClASSIFICATION

Enriching Transformers with Structured Tensor-Product Representations for Abstractive Summarization

1 code implementation NAACL 2021 Yichen Jiang, Asli Celikyilmaz, Paul Smolensky, Paul Soulos, Sudha Rao, Hamid Palangi, Roland Fernandez, Caitlin Smith, Mohit Bansal, Jianfeng Gao

On several syntactic and semantic probing tasks, we demonstrate the emergent structural information in the role vectors and improved syntactic interpretability in the TPR layer outputs.

Abstractive Text Summarization

The Out-of-Distribution Problem in Explainability and Search Methods for Feature Importance Explanations

1 code implementation NeurIPS 2021 Peter Hase, Harry Xie, Mohit Bansal

In this paper, we study several under-explored dimensions of FI explanations, providing conceptual and empirical improvements for this form of explanation.

counterfactual Feature Importance +2

DeCEMBERT: Learning from Noisy Instructional Videos via Dense Captions and Entropy Minimization

1 code implementation NAACL 2021 Zineng Tang, Jie Lei, Mohit Bansal

Second, to alleviate the temporal misalignment issue, our method incorporates an entropy minimization-based constrained attention loss, to encourage the model to automatically focus on the correct caption from a pool of candidate ASR captions.

Question Answering Retrieval +4

Cannot find the paper you are looking for? You can Submit a new open access paper.