19 citations · 64 across the 21 of their papers we have counts for
21 papers
EVPO: Explained Variance Policy Optimization for Adaptive Critic Utilization in LLM Post-Training
Chengjun Pan, Shichun Liu, Jiahang Lin +10
Reinforcement learning (RL) for LLM post-training faces a fundamental design choice: whether to use a learned critic as a baseline for policy optimization. Classical theory favors…
Inverse-Q*: Token Level Reinforcement Learning for Aligning Large Language Models Without Preference Data
Han Xia, Songyang Gao, Qiming Ge +3
Reinforcement Learning from Human Feedback (RLHF) has proven effective in aligning large language models with human intentions, yet it often relies on complex methodologies like Pr…
AgentGym: Evolving Large Language Model-based Agents across Diverse Environments
Zhiheng Xi, Yiwen Ding, Wenxiang Chen +17
Building generalist agents that can handle diverse tasks and evolve themselves across different environments is a long-term goal in the AI community. Large language models (LLMs) a…
EasyJailbreak: A Unified Framework for Jailbreaking Large Language Models
Weikang Zhou, Xiao Wang, Limao Xiong +18
Jailbreak attacks are crucial for identifying and mitigating the security vulnerabilities of Large Language Models (LLMs). They are designed to bypass safeguards and elicit prohibi…
ToolSword: Unveiling Safety Issues of Large Language Models in Tool Learning Across Three Stages
Junjie Ye, Sixian Li, Guanyu Li +6
Tool learning is widely acknowledged as a foundational approach or deploying large language models (LLMs) in real-world scenarios. While current research primarily emphasizes lever…
Navigating the OverKill in Large Language Models
Chenyu Shi, Xiao Wang, Qiming Ge +7
Large language models are meticulously aligned to be both helpful and harmless. However, recent research points to a potential overkill which means models may refuse to answer beni…
Secrets of RLHF in Large Language Models Part II: Reward Modeling
Binghai Wang, Rui Zheng, Lu Chen +24
Reinforcement Learning from Human Feedback (RLHF) has become a crucial technology for aligning language models with human values and intentions, enabling models to produce more hel…
Linear Alignment: A Closed-form Solution for Aligning Human Preferences without Tuning and Feedback
Songyang Gao, Qiming Ge, Wei Shen +9
The success of AI assistants based on Language Models (LLMs) hinges on Reinforcement Learning from Human Feedback (RLHF) to comprehend and align with user intentions. However, trad…
RoTBench: A Multi-Level Benchmark for Evaluating the Robustness of Large Language Models in Tool Learning
Junjie Ye, Yilong Wu, Songyang Gao +7
Tool learning has generated widespread interest as a vital means of interaction between Large Language Models (LLMs) and the physical world. Current research predominantly emphasiz…
ToolEyes: Fine-Grained Evaluation for Tool Learning Capabilities of Large Language Models in Real-world Scenarios
Junjie Ye, Guanyu Li, Songyang Gao +9
Existing evaluations of tool learning primarily focus on validating the alignment of selected tools for large language models (LLMs) with expected outcomes. However, these approach…
LoRAMoE: Alleviate World Knowledge Forgetting in Large Language Models via MoE-Style Plugin
Shihan Dou, Enyu Zhou, Yan Liu +13
Supervised fine-tuning (SFT) is a crucial step for large language models (LLMs), enabling them to align with human instructions and enhance their capabilities in downstream tasks.…
RealBehavior: A Framework for Faithfully Characterizing Foundation Models' Human-like Behavior Mechanisms
Enyu Zhou, Rui Zheng, Zhiheng Xi +7
Reports of human-like behaviors in foundation models are growing, with psychological theories providing enduring tools to investigate these behaviors. However, current research ten…
TRACE: A Comprehensive Benchmark for Continual Learning in Large Language Models
Xiao Wang, Yuansen Zhang, Tianze Chen +9
Aligned large language models (LLMs) demonstrate exceptional capabilities in task-solving, following instructions, and ensuring safety. However, the continual learning aspect of th…
Secrets of RLHF in Large Language Models Part I: PPO
Rui Zheng, Shihan Dou, Songyang Gao +24
Large language models (LLMs) have formulated a blueprint for the advancement of artificial general intelligence. Its primary objective is to function as a human-centric (helpful, h…
On the Universal Adversarial Perturbations for Efficient Data-free Adversarial Detection
Songyang Gao, Shihan Dou, Qi Zhang +3
Detecting adversarial samples that are carefully crafted to fool the model is a critical step to socially-secure applications. However, existing adversarial detection methods requi…
DSRM: Boost Textual Adversarial Training with Distribution Shift Risk Minimization
Songyang Gao, Shihan Dou, Yan Liu +5
Adversarial training is one of the best-performing methods in improving the robustness of deep language models. However, robust models come at the cost of high time consumption, as…
Farewell to Aimless Large-scale Pretraining: Influential Subset Selection for Language Model
Xiao Wang, Weikang Zhou, Qi Zhang +7
Pretrained language models have achieved remarkable success in various natural language processing tasks. However, pretraining has recently shifted toward larger models and larger…
CausalAPM: Generalizable Literal Disentanglement for NLU Debiasing
Songyang Gao, Shihan Dou, Junjie Shan +2
Dataset bias, i.e., the over-reliance on dataset-specific literal heuristics, is getting increasing attention for its detrimental effect on the generalization ability of NLU models…
Self-Polish: Enhance Reasoning in Large Language Models via Problem Refinement
Zhiheng Xi, Senjie Jin, Yuhao Zhou +5
To enhance the multi-step reasoning capabilities of large language models, researchers have extensively explored prompting methods, notably the Chain-of-Thought (CoT) method which…
Kernel-Whitening: Overcome Dataset Bias with Isotropic Sentence Embedding
Songyang Gao, Shihan Dou, Qi Zhang +1
Dataset bias has attracted increasing attention recently for its detrimental effect on the generalization ability of fine-tuned models. The current mainstream solution is designing…
Decorrelate Irrelevant, Purify Relevant: Overcome Textual Spurious Correlations from a Feature Perspective
Shihan Dou, Rui Zheng, Ting Wu +5
Natural language understanding (NLU) models tend to rely on spurious correlations (i.e., dataset bias) to achieve high performance on in-distribution datasets but poor performance…