Papers by Chenghao Deng

7 papers
Revealing and Enhancing Core Visual Regions: Harnessing Internal Attention Dynamics for Hallucination Mitigation in LVLMs (2026.findings-acl)

Copied to clipboard

Challenge: Existing training-free methods are vulnerable to the attention sink phenomenon . Existing methods include contrastive decoding and auxiliary expert models .
Approach: They propose a training-free attention intervention that constructs a PAD map to identify semantically core visual regions and applies per-head Median Absolute Deviation Scaling to adaptively control the intervention strength.
Outcome: The proposed intervention improves visual grounding and reduces hallucinations on multiple LVLMs and benchmarks.
Hello Again! LLM-powered Personalized Agent for Long-term Dialogue (2025.naacl-long)

Copied to clipboard

Challenge: Existing dialogue systems focus on brief single-session interactions, neglecting real-world needs for long-term companionship and personalized interactions.
Approach: They propose a model-agnostic framework for long-term dialogue agents . they use event summary and persona management to enable reasoning .
Outcome: The proposed framework incorporates three independently tunable modules dedicated to event perception, persona extraction, and response generation.
Semantically Comprehensive Token Pruning in LVLMs via Maximizing Concept Coverage (2026.acl-long)

Copied to clipboard

Challenge: Existing visual token pruning methods leverage simple metrics derived from human experience, such as attention or similarity, to rank and select tokens within a highly entangled feature space.
Approach: They propose a novel visual token pruning method that uses a concept-driven paradigm to quantify the Marginal Semantic Gain of each token's contribution to uncovered concepts.
Outcome: The proposed method outperforms state-of-the-art methods in a concept-driven model while maintaining semantic completeness.
LLM Knows Body Language, Too: Translating Speech Voices into Human Gestures (2024.acl-long)

Copied to clipboard

Challenge: despite advances in the generation of realistic human gestures, the process often includes unintended, meaningless, or non-realistic gestures.
Approach: They propose a framework that leverages large language models to generate human gestures . the primary stage employs a transformer-based auto-encoder network to encode human gesture into discrete symbols .
Outcome: The proposed framework has demonstrated state-of-the-art performance on public TED and TED-Expressive datasets.
EnsemW2S: Enhancing Weak-to-Strong Generalization with Large Language Model Ensembles (2026.findings-acl)

Copied to clipboard

Challenge: Large Language Models (LLMs) are rapidly approaching and potentially exceeding human-level performance . a novel method aims to improve weak experts' generalization abilities by training them on limited human- level data .
Approach: They propose a method that iteratively combines multiple weak experts to improve their generalization performance by training on limited human-level data.
Outcome: The proposed method improves weak experts' generalization abilities by iterating on weak models and stronger student models.
Progressive Multimodal Reasoning via Active Retrieval (2025.acl-long)

Copied to clipboard

Challenge: Existing approaches to improve multimodal large language models' reasoning performance are limited.
Approach: They propose a framework to progressively improve multimodal reasoning capabilities . they propose active retrieval and Monte Carlo tree search to improve MLLMs' reasoning .
Outcome: The proposed framework improves multimodal reasoning capabilities in multimodal large language models.
Fisher-Driven Adaptive Locating for Knowledge Editing in Large Language Models (2026.acl-long)

Copied to clipboard

Challenge: Existing methods for locating and editing static knowledge are costly and risk catastrophic forgetting or error.
Approach: They propose a Fisher-driven adaptation-aware locating strategy that dynamically identifies which model components should be edited for a given knowledge update.
Outcome: Experiments on standard benchmarks show that FiDAL improves editing effectiveness and knowledge preservation across multiple editing methods.

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