59 citations · 375 across the 88 of their papers we have counts for
44 papers · 1 filter
Every Coin Has Two Sides: On the Dual Nature of Generalization in On-Policy Distillation of Large Language Models
Zhaoyi Li, Deyang Kong, Yuan Wei +13
On-policy distillation (OPD) transfers teacher capabilities by supervising trajectories sampled from the student's own policy, yet its generalization behavior remains poorly unders…
Support Vector Rubrics: Closing the Gap Between Self-Generated and Human Rubrics
Mengyuan Sun, Yu Li, Zhuohao Yu +2
Rubric-based evaluation is a promising paradigm for judging large language model (LLM) outputs, yet self-generated rubrics lag human-annotated criteria on hard instances. We argue…
Retrieval as Generation: A Unified Framework with Self-Triggered Information Planning
Bo Li, Mingda Wang, Gexiang Fang +2
We revisit retrieval-augmented generation (RAG) by embedding retrieval control directly into generation. Instead of treating retrieval as an external intervention, we express retri…
Instruction Data Selection via Answer Divergence
Bo Li, Mingda Wang, Shikun Zhang +1
Instruction tuning relies on large instruction-response corpora whose quality and composition strongly affect downstream performance. We propose Answer Divergence-Guided Selection…
Data Selection for Multi-turn Dialogue Instruction Tuning
Bo Li, Shikun Zhang, Wei Ye
Instruction-tuned language models increasingly rely on large multi-turn dialogue corpora, but these datasets are often noisy and structurally inconsistent, with topic drift, repeti…
SteerRM: Debiasing Reward Models via Sparse Autoencoders
Mengyuan Sun, Zhuohao Yu, Weizheng Gu +2
Reward models (RMs) are critical components of alignment pipelines, yet they exhibit biases toward superficial stylistic cues, preferring better-presented responses over semantical…
Adaptive Test-Time Compute Allocation for Block Diffusion Language Models in Complex Reasoning
Yi Lu, Deyang Kong, Jianing Wang +8
Recent advances in block diffusion language models have demonstrated competitive performance and strong scalability on reasoning tasks. However, their test-time compute allocation…
ToolSafe: Enhancing Tool Invocation Safety of LLM-based agents via Proactive Step-level Guardrail and Feedback
Yutao Mou, Zhangchi Xue, Lijun Li +4
While LLM-based agents can interact with environments via invoking external tools, their expanded capabilities also amplify security risks. Monitoring step-level tool invocation be…
Modeling Uncertainty Trends for Timely Retrieval in Dynamic RAG
Bo Li, Tian Tian, Zhenghua Xu +3
Dynamic retrieval-augmented generation (RAG) allows large language models (LLMs) to fetch external knowledge on demand, offering greater adaptability than static RAG. A central cha…
Decoupling Safety into Orthogonal Subspace: Cost-Efficient and Performance-Preserving Alignment for Large Language Models
Yutao Mou, Xiaoling Zhou, Yuxiao Luo +2
Safety alignment is essential for building trustworthy artificial intelligence, yet it remains challenging to enhance model safety without degrading general performance. Current ap…
AutoRed: A Free-form Adversarial Prompt Generation Framework for Automated Red Teaming
Muxi Diao, Yutao Mou, Keqing He +6
The safety of Large Language Models (LLMs) is crucial for the development of trustworthy AI applications. Existing red teaming methods often rely on seed instructions, which limits…
Temporal Self-Rewarding Language Models: Decoupling Chosen-Rejected via Past-Future
Yidong Wang, Xin Wang, Cunxiang Wang +9
Self-Rewarding Language Models propose an architecture in which the Large Language Models(LLMs) both generates responses and evaluates its own outputs via LLM-as-a-Judge prompting,…
SAEMark: Steering Personalized Multilingual LLM Watermarks with Sparse Autoencoders
Zhuohao Yu, Xingru Jiang, Weizheng Gu +4
Watermarking LLM-generated text is critical for content attribution and misinformation prevention. However, existing methods compromise text quality, require white-box model access…
RewardAnything: Generalizable Principle-Following Reward Models
Zhuohao Yu, Jiali Zeng, Weizheng Gu +7
Reward Models, essential for guiding Large Language Model optimization, are typically trained on fixed preference datasets, resulting in rigid alignment to single, implicit prefere…
MPL: Multiple Programming Languages with Large Language Models for Information Extraction
Bo Li, Gexiang Fang, Wei Ye +4
Recent research in information extraction (IE) focuses on utilizing code-style inputs to enhance structured output generation. The intuition behind this is that the programming lan…
Can You Really Trust Code Copilots? Evaluating Large Language Models from a Code Security Perspective
Yutao Mou, Xiao Deng, Yuxiao Luo +2
Code security and usability are both essential for various coding assistant applications driven by large language models (LLMs). Current code security benchmarks focus solely on si…
SaRO: Enhancing LLM Safety through Reasoning-based Alignment
Yutao Mou, Yuxiao Luo, Shikun Zhang +1
Current safety alignment techniques for large language models (LLMs) face two key challenges: (1) under-generalization, which leaves models vulnerable to novel jailbreak attacks, a…
SampleMix: A Sample-wise Pre-training Data Mixing Strategey by Coordinating Data Quality and Diversity
Xiangyu Xi, Deyang Kong, Jian Yang +7
Existing pretraining data mixing methods for large language models (LLMs) typically follow a domain-wise methodology, a top-down process that first determines domain weights and th…
Reasoning Through Execution: Unifying Process and Outcome Rewards for Code Generation
Zhuohao Yu, Weizheng Gu, Yidong Wang +5
Large Language Models excel at code generation yet struggle with complex programming tasks that demand sophisticated reasoning. To bridge this gap, traditional process supervision…
SG-Bench: Evaluating LLM Safety Generalization Across Diverse Tasks and Prompt Types
Yutao Mou, Shikun Zhang, Wei Ye
Ensuring the safety of large language model (LLM) applications is essential for developing trustworthy artificial intelligence. Current LLM safety benchmarks have two limitations.…
RAGLAB: A Modular and Research-Oriented Unified Framework for Retrieval-Augmented Generation
Xuanwang Zhang, Yunze Song, Yidong Wang +10
Large Language Models (LLMs) demonstrate human-level capabilities in dialogue, reasoning, and knowledge retention. However, even the most advanced LLMs face challenges such as hall…
MaVEn: An Effective Multi-granularity Hybrid Visual Encoding Framework for Multimodal Large Language Model
Chaoya Jiang, Jia Hongrui, Haiyang Xu +6
This paper presents MaVEn, an innovative Multi-granularity Visual Encoding framework designed to enhance the capabilities of Multimodal Large Language Models (MLLMs) in multi-image…
Refining Corpora from a Model Calibration Perspective for Chinese Spelling Correction
Dingyao Yu, Yang An, Wei Ye +4
Chinese Spelling Correction (CSC) commonly lacks large-scale high-quality corpora, due to the labor-intensive labeling of spelling errors in real-life human writing or typing scena…
Supportiveness-based Knowledge Rewriting for Retrieval-augmented Language Modeling
Zile Qiao, Wei Ye, Yong Jiang +5
Retrieval-augmented language models (RALMs) have recently shown great potential in mitigating the limitations of implicit knowledge in LLMs, such as untimely updating of the latest…