Search Results for author:
Found 46 papers, 31 papers with code
Multi-Attribute Constraint Satisfaction via Language Model Rewriting
Thus, we create Multi-Attribute Constraint Satisfaction (MACS), a generalized method capable of finetuning language models on any sequential domain to satisfy user-specified constraints on multiple external real-value attributes.
AI as Humanity's Salieri: Quantifying Linguistic Creativity of Language Models via Systematic Attribution of Machine Text against Web Text
We present CREATIVITY INDEX as the first step to quantify the linguistic creativity of a text by reconstructing it from existing text snippets on the web.
HAICOSYSTEM: An Ecosystem for Sandboxing Safety Risks in Human-AI Interactions
AI agents are increasingly autonomous in their interactions with human users and tools, leading to increased interactional safety risks.
StyleRemix: Interpretable Authorship Obfuscation via Distillation and Perturbation of Style Elements
Authorship obfuscation, rewriting a text to intentionally obscure the identity of the author, is an important but challenging task.
Certainly Uncertain: A Benchmark and Metric for Multimodal Epistemic and Aleatoric Awareness
The ability to acknowledge the inevitable uncertainty in their knowledge and reasoning is a prerequisite for AI systems to be truly truthful and reliable.
How to Train Your Fact Verifier: Knowledge Transfer with Multimodal Open Models
Given the growing influx of misinformation across news and social media, there is a critical need for systems that can provide effective real-time verification of news claims.
WildTeaming at Scale: From In-the-Wild Jailbreaks to (Adversarially) Safer Language Models
As WildJailbreak considerably upgrades the quality and scale of existing safety resources, it uniquely enables us to examine the scaling effects of data and the interplay of data properties and model capabilities during safety training.
Information-Theoretic Distillation for Reference-less Summarization
The current winning recipe for automatic summarization is using proprietary large-scale language models (LLMs) such as ChatGPT as is, or imitation learning from them as teacher models.
JAMDEC: Unsupervised Authorship Obfuscation using Constrained Decoding over Small Language Models
The permanence of online content combined with the enhanced authorship identification techniques calls for stronger computational methods to protect the identity and privacy of online authorship when needed, e. g., blind reviews for scientific papers, anonymous online reviews, or anonymous interactions in the mental health forums.
A Roadmap to Pluralistic Alignment
We identify and formalize three possible ways to define and operationalize pluralism in AI systems: 1) Overton pluralistic models that present a spectrum of reasonable responses; 2) Steerably pluralistic models that can steer to reflect certain perspectives; and 3) Distributionally pluralistic models that are well-calibrated to a given population in distribution.
NovaCOMET: Open Commonsense Foundation Models with Symbolic Knowledge Distillation
We present NovaCOMET, an open commonsense knowledge model, that combines the best aspects of knowledge and general task models.
Localized Symbolic Knowledge Distillation for Visual Commonsense Models
Empirical results and human evaluations in a zero-shot setup demonstrate that our distillation method results in more precise VL models of reasoning compared to a baseline of passing a generated referring expression to an LLM.
The Unlocking Spell on Base LLMs: Rethinking Alignment via In-Context Learning
We analyze the effect of alignment tuning by examining the token distribution shift between base LLMs and their aligned counterpart.
STEER: Unified Style Transfer with Expert Reinforcement
We propose STEER: Unified Style Transfer with Expert Reinforcement, a unified frame-work developed to overcome the challenge of limited parallel data for style transfer.
In Search of the Long-Tail: Systematic Generation of Long-Tail Inferential Knowledge via Logical Rule Guided Search
To effectively use large language models (LLMs) for real-world queries, it is imperative that they generalize to the long-tail distribution, i. e. rare examples where models exhibit low confidence.
Tailoring Self-Rationalizers with Multi-Reward Distillation
Results on five difficult question-answering datasets StrategyQA, QuaRel, OpenBookQA, NumerSense and QASC show that not only does MaRio improve task accuracy, but it also improves the self-rationalization quality of small LMs across the aforementioned axes better than a supervised fine-tuning (SFT) baseline.
The Generative AI Paradox: "What It Can Create, It May Not Understand"
Specifically, we propose and test the Generative AI Paradox hypothesis: generative models, having been trained directly to reproduce expert-like outputs, acquire generative capabilities that are not contingent upon -- and can therefore exceed -- their ability to understand those same types of outputs.
Phenomenal Yet Puzzling: Testing Inductive Reasoning Capabilities of Language Models with Hypothesis Refinement
The ability to derive underlying principles from a handful of observations and then generalize to novel situations -- known as inductive reasoning -- is central to human intelligence.
Value Kaleidoscope: Engaging AI with Pluralistic Human Values, Rights, and Duties
To improve AI systems to better reflect value pluralism, the first-order challenge is to explore the extent to which AI systems can model pluralistic human values, rights, and duties as well as their interaction.
Faith and Fate: Limits of Transformers on Compositionality
We formulate compositional tasks as computation graphs to systematically quantify the level of complexity, and break down reasoning steps into intermediate sub-procedures.
Impossible Distillation: from Low-Quality Model to High-Quality Dataset & Model for Summarization and Paraphrasing
We present Impossible Distillation, a novel framework for paraphrasing and sentence summarization, that distills a high-quality dataset and model from a low-quality teacher that itself cannot perform these tasks.
Leftover Lunch: Advantage-based Offline Reinforcement Learning for Language Models
However, RLHF is an unstable and data-hungry process that continually requires new high-quality LM-generated data for finetuning.
Inference-Time Policy Adapters (IPA): Tailoring Extreme-Scale LMs without Fine-tuning
While extreme-scale language models have demonstrated exceptional performance on a variety of language tasks, the degree of control over these language models through pure prompting can often be limited.
Fusing Pre-Trained Language Models With Multimodal Prompts Through Reinforcement Learning
Language models are capable of commonsense reasoning: while domain-specific models can learn from explicit knowledge (e. g. commonsense graphs [6], ethical norms [25]), and larger models like GPT-3 manifest broad commonsense reasoning capacity.
SODA: Million-scale Dialogue Distillation with Social Commonsense Contextualization
Data scarcity has been a long standing issue in the field of open-domain social dialogue.
ClarifyDelphi: Reinforced Clarification Questions with Defeasibility Rewards for Social and Moral Situations
Context is everything, even in commonsense moral reasoning.
I2D2: Inductive Knowledge Distillation with NeuroLogic and Self-Imitation
Here, we investigate an alternative that a priori seems impossible: can smaller language models (e. g., GPT-2) win over models that are orders of magnitude larger and better (e. g., GPT-3), if powered with novel commonsense distillation algorithms?
Generating Sequences by Learning to Self-Correct
Sequence generation applications require satisfying semantic constraints, such as ensuring that programs are correct, using certain keywords, or avoiding undesirable content.
Rainier: Reinforced Knowledge Introspector for Commonsense Question Answering
Our work is the first to report that knowledge generated by models that are orders of magnitude smaller than GPT-3, even without direct supervision on the knowledge itself, can exceed the quality of commonsense knowledge elicited from GPT-3.
Quark: Controllable Text Generation with Reinforced Unlearning
Large-scale language models often learn behaviors that are misaligned with user expectations.
ProsocialDialog: A Prosocial Backbone for Conversational Agents
With this dataset, we introduce a dialogue safety detection module, Canary, capable of generating RoTs given conversational context, and a socially-informed dialogue agent, Prost.
Ranked #1 on
Dialogue Safety Prediction
on ProsocialDialog
Multimodal Knowledge Alignment with Reinforcement Learning
Large language models readily adapt to novel settings, even without task-specific training data.
NaturalProver: Grounded Mathematical Proof Generation with Language Models
Theorem proving in natural mathematical language - the mixture of symbolic and natural language used by humans - plays a central role in mathematical advances and education, and tests aspects of reasoning that are core to intelligence.
Twist Decoding: Diverse Generators Guide Each Other
Our extensive evaluations on machine translation and scientific paper summarization demonstrate that Twist decoding substantially outperforms each model decoded in isolation over various scenarios, including cases where domain-specific and general-purpose models are both available.
MERLOT Reserve: Neural Script Knowledge through Vision and Language and Sound
Given a video, we replace snippets of text and audio with a MASK token; the model learns by choosing the correct masked-out snippet.
Ranked #6 on
Action Classification
on Kinetics-600
(using extra training data)
NeuroLogic A*esque Decoding: Constrained Text Generation with Lookahead Heuristics
To enable constrained generation, we build on NeuroLogic decoding (Lu et al., 2021), combining its flexibility in incorporating logical constraints with A*esque estimates of future constraint satisfaction.
Ranked #1 on
Text Generation
on ROCStories
Connecting the Dots between Audio and Text without Parallel Data through Visual Knowledge Transfer
In a difficult zero-shot setting with no paired audio-text data, our model demonstrates state-of-the-art zero-shot performance on the ESC50 and US8K audio classification tasks, and even surpasses the supervised state of the art for Clotho caption retrieval (with audio queries) by 2. 2\% R@1.
Generated Knowledge Prompting for Commonsense Reasoning
It remains an open question whether incorporating external knowledge benefits commonsense reasoning while maintaining the flexibility of pretrained sequence models.
Symbolic Knowledge Distillation: from General Language Models to Commonsense Models
We apply this to the ATOMIC resource, and share our new symbolic knowledge graph and commonsense models.
MERLOT: Multimodal Neural Script Knowledge Models
As humans, we understand events in the visual world contextually, performing multimodal reasoning across time to make inferences about the past, present, and future.
DExperts: Decoding-Time Controlled Text Generation with Experts and Anti-Experts
Despite recent advances in natural language generation, it remains challenging to control attributes of generated text.
On-the-Fly Attention Modulation for Neural Generation
Despite considerable advancements with deep neural language models (LMs), neural text generation still suffers from degeneration: the generated text is repetitive, generic, self-contradictory, and often lacks commonsense.
Analyzing Commonsense Emergence in Few-shot Knowledge Models
Our results show that commonsense knowledge models can rapidly adapt from limited examples, indicating that KG fine-tuning serves to learn an interface to encoded knowledge learned during pretraining.
NeuroLogic Decoding: (Un)supervised Neural Text Generation with Predicate Logic Constraints
While the dominant recipe for conditional text generation has been large-scale pretrained language models that are finetuned on the task-specific training data, such models do not learn to follow the underlying constraints reliably, even when supervised with large amounts of task-specific examples.
Reflective Decoding: Beyond Unidirectional Generation with Off-the-Shelf Language Models
In this paper, we present Reflective Decoding, a novel unsupervised algorithm that allows for direct application of unidirectional LMs to non-sequential tasks.
HATNet: An End-to-End Holistic Attention Network for Diagnosis of Breast Biopsy Images
HATNet extends the bag-of-words approach and uses self-attention to encode global information, allowing it to learn representations from clinically relevant tissue structures without any explicit supervision.
