Papers by Zhenglin Wang
Safety Sidecar: Reflection-Driven Runtime Control for Safer Agents (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing safety controls fail to provide runtime intervention or cross-architecture portability for autonomous LLM agents. |
| Approach: | They propose a model-agnostic, plug-and-play module to provide arbitrary agent safety control and auditability. |
| Outcome: | The proposed module improves the secure-solution rate by 2.9–11.2 percentage points . it adds only 3.2s to end-to-end latency and a negligible average cost of 5.37 10-4 per scenario . |
AdaRewriter: Unleashing the Power of Prompting-based Conversational Query Reformulation via Test-Time Adaptation (2025.emnlp-main)
Copied to clipboard
| Challenge: | Prompting-based conversational query reformulation has emerged as a powerful approach for conversational search, refining ambiguous user queries into standalone search queries. |
| Approach: | They propose a framework for query reformulation using an outcome-supervised reward model via test-time adaptation. |
| Outcome: | Experiments on five conversational search datasets show that AdaRewriter significantly outperforms the existing methods across most settings. |
SCOPE: Optimizing Key-Value Cache Compression in Long-context Generation (2025.acl-long)
Copied to clipboard
| Challenge: | Excessive compression during the prefill phase impairs comprehension of reasoning tasks . SCOPE is a framework that performs KV cache optimization during the decoding and prefill phases . |
| Approach: | They propose a framework that performs optimization during the prefill and decoding phases . they propose enabling a sliding strategy to select essential heavy hitters for the decoding phase . |
| Outcome: | Experiments show that SCOPE can optimize key-value cache for long-context generation tasks . the framework can preserve essential information while minimizing memory usage and transfer . |
VReST: Enhancing Reasoning in Large Vision-Language Models through Tree Search and Self-Reward Mechanism (2025.acl-long)
Copied to clipboard
| Challenge: | Large Vision-Language Models (LVLMs) have shown exceptional performance in multimodal tasks, but their effectiveness in complex visual reasoning is constrained. |
| Approach: | They propose a training-free approach that enhances Reasoning in Large Vision-Language Models . they propose integrating Monte Carlo Tree Search and Self-Reward mechanisms into the reasoning tree . |
| Outcome: | The proposed approach surpasses current prompting methods and secures state-of-the-art performance across three multimodal reasoning benchmarks. |
WebWalker: Benchmarking LLMs in Web Traversal (2025.acl-long)
Copied to clipboard
Jialong Wu, Wenbiao Yin, Yong Jiang, Zhenglin Wang, Zekun Xi, Runnan Fang, Linhai Zhang, Yulan He, Deyu Zhou, Pengjun Xie, Fei Huang
| Challenge: | Large Language Models (LLMs) have demonstrated impressive capabilities across a wide range of natural language processing tasks. |
| Approach: | They propose a benchmark to assess the ability of LLMs to perform web traversal by using an explore-critic paradigm. |
| Outcome: | The proposed framework mimics human-like web navigation through an explore-critic paradigm and demonstrates the effectiveness of RAG combined with WebWalker in real-world scenarios. |
Cracking the Code of Hallucination in LVLMs with Vision-aware Head Divergence (2025.acl-long)
Copied to clipboard
Jinghan He, Kuan Zhu, Haiyun Guo, Junfeng Fang, Zhenglin Hua, Yuheng Jia, Ming Tang, Tat-Seng Chua, Jinqiao Wang
| Challenge: | Existing methods focus on alignment training or decoding refinements but address symptoms at the generation stage without probing the underlying causes. |
| Approach: | They propose a training-free approach to mitigate hallucination by enhancing the role of vision-aware attention heads. |
| Outcome: | The proposed method achieves superior performance compared to state-of-the-art approaches in mitigating hallucinations while maintaining high efficiency with negligible additional time overhead. |
When KV Cache Reuse Fails in Multi-Agent Systems: Cross-Candidate Interaction is Crucial for LLM Judges (2026.acl-long)
Copied to clipboard
| Challenge: | Multi-agent LLMs generate multiple candidate responses that are aggregated by an LLM judge. |
| Approach: | They propose to advocate KV cache reuse across partially shared contexts and report substantial speedups for generation agents. |
| Outcome: | The proposed reuse strategies weaken cross-candidate attention, especially for later candidate blocks, and highlight judge-centric inference as a distinct regime that requires dedicated, risk-aware system design. |
ConnPrompt: Connective-cloze Prompt Learning for Implicit Discourse Relation Recognition (2022.coling-1)
Copied to clipboard
| Challenge: | Existing paradigms for Implicit Discourse Relation Recognition (IDRR) do not exploit linguistic evidence embedded in the pre-training process. |
| Approach: | They propose a new paradigm to detect and classify relation sense between two text segments without an explicit connective. |
| Outcome: | The proposed method significantly outperforms the state-of-the-art algorithms even with fewer training data. |
Large Language Models Have Intrinsic Meta-Cognition, but Need a Good Lens (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing studies have focused on the cognitive error detection capabilities of Large Language Models (LLMs), but few studies have examined the meta-cognitive abilities of LLMs. |
| Approach: | They propose an automated meta-cognition evaluation framework for evaluation of LLMs and a Markovian Intrinsic Reward Adjustment strategy to boost current lenses. |
| Outcome: | The proposed framework can be used to evaluate the meta-cognition abilities of LLMs and improve them. |
SEED: Accelerating Reasoning Tree Construction via Scheduled Speculative Decoding (2025.coling-main)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have remarkable emergent abilities across various tasks, yet their performance on complex reasoning and planning tasks remains suboptimal. |
| Approach: | They propose a tree-search-based reasoning framework that encourages the exploration of intermediate steps and a round-scheduled strategy to manage draft model dispatching. |
| Outcome: | The proposed framework improves runtime speed and GPU memory management concurrently and handles multiple iterations for thought generation and state evaluation. |
Opinions Are Not Always Positive: Debiasing Opinion Summarization with Model-Specific and Model-Agnostic Methods (2024.lrec-main)
Copied to clipboard
| Challenge: | Existing opinion summarization frameworks are reluctant to generate negative summaries given input of negative opinions. |
| Approach: | They propose to disentangle input into sentiment-relevant and sentiment-irrelevant components through adversarial loss. |
| Outcome: | The proposed approaches reduce sentiment bias in the existing opinion summarization dataset . the proposed approaches generate better summaries with a more balanced emotional polarity distribution . |
Multimodal Self-Instruct: Synthetic Abstract Image and Visual Reasoning Instruction Using Language Model (2024.emnlp-main)
Copied to clipboard
Wenqi Zhang, Zhenglin Cheng, Yuanyu He, Mengna Wang, Yongliang Shen, Zeqi Tan, Guiyang Hou, Mingqian He, Yanna Ma, Weiming Lu, Yueting Zhuang
| Challenge: | Using large language models, large multimodal models struggle with basic tasks like reading time from a clock and planning a route using a road map. |
| Approach: | They propose a multimodal self-instruct that synthesizes massive abstract images and visual reasoning instructions. |
| Outcome: | The proposed model synthesizes 11,193 abstract images and reasoning instructions across eight visual scenarios. |
Benchmarking Temporal Reasoning and Alignment Across Chinese Dynasties (2026.eacl-short)
Copied to clipboard
| Challenge: | Existing temporal reasoning benchmarks rely on rule-based construction and lack contextual depth . a recent study found existing LLMs struggle with nuanced temporal understanding . |
| Approach: | a benchmark is designed to evaluate LLMs on temporal reasoning in Chinese dynasties. |
| Outcome: | a new benchmark evaluates LLMs on temporal reasoning across Chinese dynasties . it emphasizes cross-entity relationships, pairwise temporal alignment, contextualized and culturally-grounded reasoning . results show existing LLM benchmarks struggle with nuanced temporal understanding . |