Papers by Yujun Zhang

11 papers
SceMQA: A Scientific College Entrance Level Multimodal Question Answering Benchmark (2024.acl-short)

Copied to clipboard

Challenge: SceMQA focuses on core science subjects including Mathematics, Physics, Chemistry, and Biology.
Approach: They propose to use SceMQA to evaluate multimodal question answering at college entrance level.
Outcome: The proposed model provides specific knowledge points for each problem and detailed explanations for each answer.
Too Correct to Learn: Reinforcement Learning on Saturated Reasoning Data (2026.acl-short)

Copied to clipboard

Challenge: Strong base models saturate benchmarks, resulting in weaker performance, a paradox . a new approach to Reinforcement Learning (RL) is needed to improve performance .
Approach: They propose a method that uses constrained uniform top-k sampling to flatten the local optimization landscape by sampling uniformly from constrained high-confidence candidates.
Outcome: Experiments show that the proposed approach prevents policy degeneration and boosts out-of-domain generalization.
Beyond Single-Value Metrics: Evaluating and Enhancing LLM Unlearning with Cognitive Diagnosis (2025.findings-acl)

Copied to clipboard

Challenge: Large Language Models (LLMs) have been used to remove harmful knowledge and undesirable capabilities.
Approach: They propose a framework that leverages Cognitive Diagnosis Modeling to evaluate LLM unlearning.
Outcome: The proposed framework enhances evaluation and facilitates removal of harmful abilities.
A Structure-Aware Argument Encoder for Literature Discourse Analysis (2022.coling-1)

Copied to clipboard

Challenge: Existing research for argument representation learning treats tokens in sentences equally and ignores the implied structure information of argumentative context.
Approach: They propose to separate tokens into two groups to capture structural information of arguments and to incorporate paragraph-level position information into the model.
Outcome: The proposed model captures structural information of arguments and is able to identify arguments automatically.
Dynamic Infilling Anchors for Format-Constrained Generation in Diffusion Large Language Models (2026.acl-long)

Copied to clipboard

Challenge: Diffusion large language models (dLLMs) offer bidirectional attention and parallel generation . fixed anchors can enforce constraints, but they often impose rigid spans, leading to truncated reasoning .
Approach: They propose a method that dynamically estimates end-anchor positions to adjust generation length before iterative infilling.
Outcome: The proposed method improves format compliance and answer accuracy on GSM8K and MATH.
PolicyLLM: Towards Excellent Comprehension of Public Policy for Large Language Models (2026.findings-acl)

Copied to clipboard

Challenge: Large Language Models (LLMs) are increasingly integrated into real-world decision-making, but their ability to comprehend and reason about policy-related content remains underexplored.
Approach: They propose a bilingual benchmark evaluating policy comprehension comprising 21K cases across a broad spectrum of policy areas.
Outcome: The proposed model shows stronger performance on application-oriented policy tasks than on memorization or conceptual understanding, and yields the highest accuracy on structured reasoning tasks.
Judging with Many Minds: Do More Perspectives Mean Less Prejudice? On Bias Amplification and Resistance in Multi-Agent Based LLM-as-Judge (2025.findings-emnlp)

Copied to clipboard

Challenge: LLM-as-Judge frameworks provide scalable alternative to human evaluation . but the question of how intrinsic biases manifest in these settings remains unexplored .
Approach: They conduct systematic analysis of four bias types in multi-agent LLM-as-Judge frameworks . they find debate framework amplifies biases sharply after initial debate .
Outcome: The proposed frameworks amplify biases after debate and show they are stronger in meta-judge scenarios.
Dissecting Logical Reasoning in LLMs: A Fine-Grained Evaluation and Supervision Study (2025.findings-emnlp)

Copied to clipboard

Challenge: Existing benchmarks that rely on final-answer accuracy fail to capture the quality of the reasoning process.
Approach: They propose a fine-grained evaluation framework that assesses logical reasoning across three dimensions: overall accuracy, stepwise soundness, and representation-level probing.
Outcome: The proposed framework assesses logical reasoning across three dimensions: overall accuracy, stepwise soundness, and representation-level probing.
Defending Jailbreak Prompts via In-Context Adversarial Game (2024.emnlp-main)

Copied to clipboard

Challenge: Large Language Models (LLMs) demonstrate remarkable capabilities across diverse applications, but concerns regarding their security persist.
Approach: They propose an adversarial game that leverages agent learning to extend knowledge to defend against jailbreaks.
Outcome: The proposed game shows that LLMs safeguarded by ICAG exhibit significantly reduced jailbreak success rates across various attack scenarios.
Graph-Assisted Large Language Models: A Perspective on Mitigating Intrinsic Limitations (2026.findings-acl)

Copied to clipboard

Challenge: Large language models exhibit intrinsic limitations such as knowledge cutoff, single-threaded reasoning that hinders finer-grained branch and aggregation, and rigid collaboration mechanisms that struggle to coordinate specialized capabilities.
Approach: They propose a taxonomy spanning *Graph-Assisted Knowledge Augmentation*, *Graph Assisted Reasoning and Planning*, and *Graphed LLM Collaboration*.
Outcome: The proposed models show that graphs can augment and correct LLMs and support dynamic coordination among experts and agents in collaborative settings.
VideoStir: Understanding Long Videos via Spatio-Temporally Structured and Intent-Aware RAG (2026.acl-long)

Copied to clipboard

Challenge: Existing methods for retrieval-augmented generation (RAG) to long videos are limited by limited context windows and flatten videos into independent segments.
Approach: They propose a structured and intent-aware long-video RAG framework that structures a video as a spatio-temporal graph and then performs multi-hop retrieval to aggregate evidence across distant yet contextually related events.
Outcome: The proposed framework is competitive with state-of-the-art baselines without auxiliary information.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations