Search Results for author:
Found 29 papers, 14 papers with code
Learning to Reason via Mixture-of-Thought for Logical Reasoning
To fill in this gap, we propose Mixture-of-Thought (MoT), a framework that enables LLMs to reason across three complementary modalities: natural language, code, and a newly introduced symbolic modality, truth-table, which systematically enumerates logical cases and partially mitigates key failure modes in natural language reasoning.
Creativity or Brute Force? Using Brainteasers as a Window into the Problem-Solving Abilities of Large Language Models
Accuracy remains a standard metric for evaluating AI systems, but it offers limited insight into how models arrive at their solutions.
Self-rewarding correction for mathematical reasoning
We study self-rewarding reasoning large language models (LLMs), which can simultaneously generate step-by-step reasoning and evaluate the correctness of their outputs during the inference time-without external feedback.
OmnixR: Evaluating Omni-modality Language Models on Reasoning across Modalities
To address this, OmnixR offers two evaluation variants: (1)synthetic subset: a synthetic dataset generated automatically by translating text into multiple modalities--audio, images, video, and hybrids (Omnify).
RRM: Robust Reward Model Training Mitigates Reward Hacking
Our RRM improves the performance of a pairwise reward model trained on Gemma-2-9b-it, on RewardBench, increasing accuracy from 80. 61% to 84. 15%.
From Lists to Emojis: How Format Bias Affects Model Alignment
In this work, we extend the study of biases in preference learning beyond the commonly recognized length bias, offering a comprehensive analysis of a wider range of format biases.
OPTune: Efficient Online Preference Tuning
In this paper, we propose a more efficient data exploration strategy for online preference tuning (OPTune), which does not rely on human-curated or pre-collected teacher responses but dynamically samples informative responses for on-policy preference alignment.
Unveiling the Impact of Coding Data Instruction Fine-Tuning on Large Language Models Reasoning
First, coding data tuning enhances the overall reasoning capabilities of LLMs across different model families and scales.
Spectrum AUC Difference (SAUCD): Human-aligned 3D Shape Evaluation
Then, we calculate the Area Under the Curve (AUC) difference between the two spectrums, so that each frequency band that captures either the overall or detailed shape is equitably considered.
Your Vision-Language Model Itself Is a Strong Filter: Towards High-Quality Instruction Tuning with Data Selection
In the first stage, we devise a scoring network to evaluate the difficulty of training instructions, which is co-trained with the VLM.
Can LLMs Speak For Diverse People? Tuning LLMs via Debate to Generate Controllable Controversial Statements
To examine DEBATUNE, we curate the largest dataset of debate topics so far, which covers 710 controversial topics and corresponding arguments for each topic.
Selective Reflection-Tuning: Student-Selected Data Recycling for LLM Instruction-Tuning
This paper introduces Selective Reflection-Tuning, a novel paradigm that synergizes a teacher LLM's reflection and introspection for improving existing data quality with the data selection capability of the student LLM, to automatically refine existing instruction-tuning data.
ODIN: Disentangled Reward Mitigates Hacking in RLHF
In this work, we study the issue of reward hacking on the response length, a challenge emerging in Reinforcement Learning from Human Feedback (RLHF) on LLMs.
GPT-4 Vision on Medical Image Classification -- A Case Study on COVID-19 Dataset
This technical report delves into the application of GPT-4 Vision (GPT-4V) in the nuanced realm of COVID-19 image classification, leveraging the transformative potential of in-context learning to enhance diagnostic processes.
HallusionBench: An Advanced Diagnostic Suite for Entangled Language Hallucination and Visual Illusion in Large Vision-Language Models
Our comprehensive case studies within HallusionBench shed light on the challenges of hallucination and illusion in LVLMs.
Ranked #1 on
Visual Question Answering (VQA)
on HallusionBench
AlpaCare:Instruction-tuned Large Language Models for Medical Application
Instruction-finetuning (IFT) has become crucial in aligning Large Language Models (LLMs) with diverse human needs and has shown great potential in medical applications.
Reflection-Tuning: Data Recycling Improves LLM Instruction-Tuning
Recent advancements in Large Language Models (LLMs) have expanded the horizons of natural language understanding and generation.
From Quantity to Quality: Boosting LLM Performance with Self-Guided Data Selection for Instruction Tuning
In the realm of Large Language Models (LLMs), the balance between instruction data quality and quantity is a focal point.
Backdooring Instruction-Tuned Large Language Models with Virtual Prompt Injection
To demonstrate the threat, we propose a simple method to perform VPI by poisoning the model's instruction tuning data, which proves highly effective in steering the LLM.
AlpaGasus: Training A Better Alpaca with Fewer Data
Large language models (LLMs) strengthen instruction-following capability through instruction-finetuning (IFT) on supervised instruction/response data.
InstructZero: Efficient Instruction Optimization for Black-Box Large Language Models
Large language models~(LLMs) are instruction followers, but it can be challenging to find the best instruction for different situations, especially for black-box LLMs on which backpropagation is forbidden.
Prompting Language-Informed Distribution for Compositional Zero-Shot Learning
Compositional zero-shot learning (CZSL) task aims to recognize unseen compositional visual concepts, e. g., sliced tomatoes, where the model is learned only from the seen compositions, e. g., sliced potatoes and red tomatoes.
Backdoor Learning on Sequence to Sequence Models
Backdoor learning has become an emerging research area towards building a trustworthy machine learning system.
PTP: Boosting Stability and Performance of Prompt Tuning with Perturbation-Based Regularizer
To address this critical problem, we first investigate and find that the loss landscape of vanilla prompt tuning is precipitous when it is visualized, where a slight change of input data can cause a big fluctuation in the loss landscape.
Dynamic Voxel Grid Optimization for High-Fidelity RGB-D Supervised Surface Reconstruction
Direct optimization of interpolated features on multi-resolution voxel grids has emerged as a more efficient alternative to MLP-like modules.
When do you need Chain-of-Thought Prompting for ChatGPT?
However, it is not clear whether CoT is still effective on more recent instruction finetuned (IFT) LLMs such as ChatGPT.
How Many Demonstrations Do You Need for In-context Learning?
Moreover, ICL (with and w/o CoT) using only one correct demo significantly outperforms all-demo ICL adopted by most previous works, indicating the weakness of LLMs in finding correct demo(s) for input queries, which is difficult to evaluate on the biased datasets.
Task-Aware Sampling Layer for Point-Wise Analysis
Sampling, grouping, and aggregation are three important components in the multi-scale analysis of point clouds.
Graph Edit Distance Reward: Learning to Edit Scene Graph
Scene Graph, as a vital tool to bridge the gap between language domain and image domain, has been widely adopted in the cross-modality task like VQA.
