Yue Zhang

Full Professor@Westlake University·中国·OpenReview

研究方向

LLM · OOD Generalization · Reasoning · Summarization · Semantic parsing · Information Extraction · Generation · Machine Translation · Sentiment · Syntactic Parsing

Yue Zhang

TrustJudge: Inconsistencies of LLM-as-a-Judge and How to Alleviate Them

Beyond English-Centric Training: How Reinforcement Learning Improves Cross-Lingual Reasoning in LLMs

Video Summarization Pretraining with Self-Discovery of Informative Frames

Video Language Models are Human-Aligned Evaluators for Text to Motion Generation

FiRE: Fine-Grained Ranking Evaluation for Machine Translation

SafeReview: Building a Robust Deep Review Assistant Against Prompt Injection

Deep Literature Survey Automation with an Iterative Workflow

DeepScientist: Advancing Frontier-Pushing Scientific Findings Progressively

RewardAnything: Generalizable Principle-Following Reward Models

AutoFigure: Generating and Refining Publication-Ready Scientific Illustrations

Safetylock: Guarding LLM againt FuneTuning Risks with Efficient Inference-time Addon

Res-Bench: Reasoning Skill-Aware Reasoning Diagnostic evaluation benchmark for math reasoning

League: Leaderboard Generation on Demand

Evaluating the Logical Reasoning Abilities of Large Reasoning Models

MMQA: Evaluating LLMs with Multi-Table Multi-Hop Complex Questions

Learning to Reason under Off-Policy Guidance

Task Calibration: Calibrating Large Language Models on Inference Tasks

ELICIT: LLM Augmentation Via External In-context Capability

CycleResearcher: Improving Automated Research via Automated Review

NovelQA: Benchmarking Question Answering on Documents Exceeding 200K Tokens

Glimpse: Enabling White-Box Methods to Use Proprietary Models for Zero-Shot LLM-Generated Text Detection

Personality Alignment of Large Language Models

Towards Homogeneous Lexical Tone Decoding from Heterogeneous Intracranial Recordings

CofCA: A STEP-WISE Counterfactual Multi-hop QA benchmark

Learning to Rank for In-Context Example Retrieval

An Empirical Analysis of Uncertainty in Large Language Model Evaluations

Direct Preference Optimization Using Sparse Feature-level Constraints

Human Simulacra: Benchmarking the Personification of Large Language Models

Constrain Alignment with Sparse Autoencoders

BALCONI: BALancing CONtext and Internal Knowledge For Training Flexible LLMs

Locking Down the Finetuned LLMs Safety

Keys to Robust Edits: From Theoretical Insights to Practical Advances

Logic Agent: Enhancing Validity with Logic Rule Invocation

An Efficient Quantum Classifier Based on Hamiltonian Representations