Papers by Lei Zhong

29 papers
MentalSeek-Dx: Towards Progressive Hypothetico-Deductive Reasoning for Real-world Psychiatric Diagnosis (2026.acl-long)

Copied to clipboard

Challenge: Mental health disorders represent a burgeoning global public health challenge . lack of ecological validity and fine-grained diagnostic supervision limits their utility .
Approach: They propose a medical-specialized LLM trained to internalize clinical reasoning process through supervised trajectory construction and curriculum-based reinforcement learning.
Outcome: The proposed model achieves state-of-the-art with only 14B parameters, establishing a clinically grounded framework for reliable psychiatric diagnosis.
Chart-MRAG: Benchmarking Multimodal Retrieval Augmented Generation on Chart-based Documents (2026.acl-long)

Copied to clipboard

Challenge: Existing benchmarks focus on simple image-text interactions, overlooking complex visual formats like charts.
Approach: They propose a semi-automatic framework for generating evaluation samples through multi-modal keypoint extraction, knowledge graph construction, and qa pair synthesis.
Outcome: The proposed framework generates 4,738 question-answering pairs across 8 domains from real-world documents.
HSUGA: LLM-Enhanced Recommendation with Hierarchical Semantic Understanding and Group-Aware Alignment (2026.findings-acl)

Copied to clipboard

Challenge: Existing methods for enhancing sequential recommendation use long interaction sequences, but they lack the ability to extract user preferences from long sequences.
Approach: They propose a plugin that integrates LLMs to infer user preferences from interaction sequences.
Outcome: The proposed algorithms improve user semantic embedding extraction and utilization on three benchmark datasets.
Article Reranking by Memory-Enhanced Key Sentence Matching for Detecting Previously Fact-Checked Claims (2021.acl-long)

Copied to clipboard

Challenge: Existing methods to detect false claims ignore the characteristics of FC-articles . claims are often quoted to describe checked events, providing lexical information . sentence templates to introduce or debunk claims are common across articles, providing pattern information.
Approach: They propose a model to rerank FC-articles using key sentences and pattern information.
Outcome: The proposed model outperforms existing methods on two real-world datasets showing that key sentences can be used to predict if an article fact-checks the given claim.
GEM: A General Evaluation Benchmark for Multimodal Tasks (2021.findings-acl)

Copied to clipboard

Challenge: Existing datasets that focus on natural language tasks are not considered as a general evaluation benchmark for multimodal tasks.
Approach: They present a general evaluation benchmark for multimodal tasks, GEM 1 . they compare it with existing multimodal vision-language datasets .
Outcome: The proposed model is compared with existing vision-language datasets focusing on natural language tasks . it is the largest vision-linguistic dataset covering image-language tasks and video-language task at the same time .
Alleviating Hallucinations from Knowledge Misalignment in Large Language Models via Selective Abstention Learning (2025.acl-long)

Copied to clipboard

Challenge: Large language models (LLMs) suffer from severe hallucination issues due to the knowledge misalignment between the pre-training stage and the supervised fine-tuning stage.
Approach: They propose a training objective with an abstention mechanism that selectively rejects tokens that misalign with the desired knowledge distribution via a special [REJ] token.
Outcome: The proposed model selectively rejects tokens that misalign with the desired knowledge distribution via a special [REJ] token.
CONE: An Efficient COarse-to-fiNE Alignment Framework for Long Video Temporal Grounding (2023.acl-long)

Copied to clipboard

Challenge: Existing work on video temporal grounding for long videos is limited by existing datasets.
Approach: They propose a query-guided window selection strategy and a coarse-to-fine mechanism to speed up inference for long videos.
Outcome: The proposed framework accelerates inference time by 2x on Ego4D-NLQ and 15x on MAD while keeping SOTA results.
Think-Search-Patch: A Retrieval-Augmented Reasoning Framework for Repository-Level Code Repair (2025.emnlp-industry)

Copied to clipboard

Challenge: Large language models suffer from multiple-file coding scenarios with strong inter-file dependencies . experimental results show that large language models exhibit inadequate performance in multi-file scenarios .
Approach: They propose a retrieval-augmented reasoning framework for repository-level code repair . they use a dataset to generate standardized patches based on the key snippets .
Outcome: The proposed framework improves retrieval accuracy and repair success on SWE-bench Lite . it surpasses models with larger size in managing extensive code contexts and fixing bugs spanning across multiple files.
Detecting Hallucinations in Retrieval-Augmented Generation via Semantic-level Internal Reasoning Graph (2026.findings-acl)

Copied to clipboard

Challenge: Existing methods for detecting faithfulness hallucinations are coarse or do not capture the models’ internal reasoning processes, making it difficult to learn.
Approach: They propose a semantic-level internal reasoning graph-based method for detecting faithfulness hallucination using Large language models.
Outcome: The proposed method achieves better overall performance compared to state-of-the-art baselines on RAGTruth and Dolly-15k.
Investigating and Mitigating the Multimodal Hallucination Snowballing in Large Vision-Language Models (2024.acl-long)

Copied to clipboard

Challenge: Large Vision-Language Models (LVLMs) suffer from multimodal hallucinations . however, the generated hallucines could influence the models’ subsequent generation .
Approach: They propose a framework to evaluate LVLMs' behaviors when encountering generated hallucinations and a method to revise the output distribution of LVLs with the one derived from the residual visual input.
Outcome: The proposed framework reduces the performance of open-source LVLMs by 31%, indicating that they are prone to accept the generated hallucinations and make false claims that they would not have supported without distractions.
MOOCCube: A Large-scale Data Repository for NLP Applications in MOOCs (2020.acl-main)

Copied to clipboard

Challenge: Massive open online courses (MOOCs) are a popular educational platform for advanced research.
Approach: They propose to use MOOCCube to build a large-scale data repository of over 700 MOOC courses, 100k concepts, 8 million student behaviors with an external resource.
Outcome: The proposed datasets show that they can facilitate research in MOOCs.
Long Chain-of-Thought Fine-tuning via Understanding-to-Reasoning Transition (2025.emnlp-main)

Copied to clipboard

Challenge: Existing research on long-context scaling in language models has focused on managing lengthy input prompts instead of producing long outputs.
Approach: They propose a sequence-level curriculum learning framework that shifts a model’s focus from interpreting long chain-of-thoughts to generating them.
Outcome: Experiments on rigorous reasoning benchmarks, including AIME24 and GPQA Diamond, show that the proposed approach surpasses standard fine-tuning by over 10% while maintaining robust performance on understanding tasks.
Latent Distribution Decouple for Uncertain-Aware Multimodal Multi-label Emotion Recognition (2025.findings-acl)

Copied to clipboard

Challenge: Existing studies focus on improving fusion strategies and modeling modality-to-label dependencies, but they overlook the impact of aleatoric uncertainty, which is inherent noise in multimodal data.
Approach: They propose a latent emotional distribution decomposition with uncertainty perception framework to model aleatoric uncertainty in multimodal data.
Outcome: The proposed framework achieves state-of-the-art performance on the CMU-MOSEI and M3ED datasets, highlighting the importance of uncertainty modeling in MMER.
CausalGaze: Unveiling Hallucinations via Counterfactual Graph Intervention in Large Language Models (2026.findings-acl)

Copied to clipboard

Challenge: Existing classification-based methods capture noise and spurious correlations while overlooking the underlying causal mechanisms.
Approach: They propose a hallucination detection framework based on structural causal models that captures static and passive signals from internal states and employs counterfactual interventions to disentangle causal reasoning paths from incidental noise.
Outcome: Experiments on four datasets and three widely used LLMs show that the proposed framework improves AUROC and interpretability.
Augmenting Legal Judgment Prediction with Contrastive Case Relations (2022.coling-1)

Copied to clipboard

Challenge: Existing legal judgment prediction methods only consider one case fact description as input, which may not fully utilize information in the data such as case relations and frequency.
Approach: They propose a new perspective that introduces some contrastive case relations to construct case triples as input and a corresponding judgment prediction framework with case triple modeling.
Outcome: The proposed framework can be used to refine encoding and decoding processes using three customized modules on two public datasets.
DecEx-RAG: Boosting Agentic Retrieval-Augmented Generation with Decision and Execution Optimization via Process Supervision (2025.emnlp-industry)

Copied to clipboard

Challenge: Recent advances in outcome-supervised reinforcement learning (RL) have shown strong performance, but this approach still suffers from inefficient exploration, sparse reward signals, and ambiguous global reward feedback.
Approach: They propose a model that models RAG as a Markov Decision Process (MDP) and introduces an efficient pruning strategy to optimize data expansion.
Outcome: The proposed model outperforms existing methods and achieves an average performance improvement of 6.2% across six datasets.
Question Tells You Where the Answer Is: Intention-aware Long-Context KV Cache Compression (2026.acl-long)

Copied to clipboard

Challenge: Recent methods to reduce the KV cache size fail to identify crucial KVs for generation while excluding others accurately, resulting in severe information loss.
Approach: They propose an intention-aware KV cache eviction method that identifies and retains crucial KVs according to the attention distribution of intention, which semantically reflects the user’s goal and determines which part of the context is relevant.
Outcome: The proposed method can maintain the model performance while reducing the KV cache size from 128K to 2K, leading to a 6.3x increase in decoding speed and 7.8x enhancement in memory efficiency compared to the default setting.
Understanding the RoPE Extensions of Long-Context LLMs: An Attention Perspective (2025.coling-main)

Copied to clipboard

Challenge: Enabling LLMs to handle lengthy context is currently a research hotspot . a notable challenge limiting further customization is the inability of LLM to utilize context beyond pretrained length due to the inherent flaw of rotary position embedding (RoPE).
Approach: They propose to extend the RoPE from an attention perspective and on two benchmarking tasks.
Outcome: The proposed extension of the RoPE improves extrapolation and retrieval errors.
Length Controlled Generation for Black-box LLMs (2025.acl-long)

Copied to clipboard

Challenge: Existing length control methods involve fine-tuning the parameters of LLMs, which is inefficient and suboptimal for practical use.
Approach: They propose an iterative sampling framework that regulates LLMs to generate length-constrained text without modifying the underlying parameters.
Outcome: The proposed method achieves 100% success rates on Llama3.1 tasks with minimal additional computational overhead.
GroupRank: A Groupwise Paradigm for Effective and Efficient Passage Reranking with LLMs (2026.findings-acl)

Copied to clipboard

Challenge: Existing rankers excel in lexical-matching scenarios, while they struggle with complex queries requiring deep reasoning.
Approach: They propose a new paradigm that balances flexibility and context awareness to unlock the full potential of groupwise reranking.
Outcome: The proposed approach achieves a state-of-the-art 65.2 NDCG@10 on BRIGHT and surpasses baselines by 2.1 points on R2MED while delivering a 6.4 inference speedup.
Training Language Models with Memory Augmentation (2022.emnlp-main)

Copied to clipboard

Challenge: Existing methods for training memory-augmented language models only introduce mem-ories at testing time or represent them using a separately trained encoder.
Approach: They propose a training approach that directly takes in-batch examples as accessible memory and new methods for memory construction and data batching that are used for adapting to different sets of memories at testing time.
Outcome: The proposed approach reduces perplexity from 18.70 to 15.37 on multiple language modeling and machine translation benchmarks.
Integrating Semantic and Structural Information with Graph Convolutional Network for Controversy Detection (2020.acl-main)

Copied to clipboard

Challenge: Existing methods for identifying controversial posts on social media are limited . existing methods fail to incorporate semantic information from content-related posts .
Approach: They propose a method to integrate the information from topics, posts, and comments . they extend their model to Disentangled TPC-GCN to disentangle topic-related features .
Outcome: The proposed method outperforms existing methods on two real-world datasets.
ZigZagKV: Dynamic KV Cache Compression for Long-context Modeling based on Layer Uncertainty (2025.coling-main)

Copied to clipboard

Challenge: Existing methods to accelerate inference of Large Language models (LLMs) are limited in their ability to retain key tokens as input length increases.
Approach: They propose a method that leverages layer uncertainty to allocate budget size for each layer to reduce memory usage.
Outcome: The proposed method reduces memory usage of the KV caches to only 20% when compared to full KV inference while achieving nearly lossless performance.
ImgTrojan: Jailbreaking Vision-Language Models with ONE Image (2025.naacl-long)

Copied to clipboard

Challenge: Existing studies on the safety of large language models (LLMs) with human values have focused on the integration of multi-modal user input into these models.
Approach: They propose a method to bypass safety constraints of large language models by using poisoned images instead of original textual captions.
Outcome: The proposed attack bypasses safety constraints of large language models (VLMs) by replacing the original textual captions with malicious jailbreak prompts.
Advancing Large Language Model Attribution through Self-Improving (2024.emnlp-main)

Copied to clipboard

Challenge: Teaching large language models to generate text with citations to evidence sources requires high-quality attribution data, which is costly and labor-intensive.
Approach: They propose a framework for iteratively improving the attribution capability of large language models (LLMs) by attributing output to verifiable sources.
Outcome: Experiments on three open-domain question-answering datasets show that START improves in aggregating information across multiple sources.
Unveiling Entity-Level Unlearning for Large Language Models: A Comprehensive Analysis (2025.coling-main)

Copied to clipboard

Challenge: Existing studies have focused on instance-level unlearning, specifically removing predefined instances containing sensitive content.
Approach: They propose a task to erase entity-related knowledge from the target model completely by analyzing the forget set and its size.
Outcome: The proposed task systematically evaluates popular unlearning algorithms and reveals that the knowledge coverage of the forget set and its size play pivotal roles.
Learning Fine-Grained Grounded Citations for Attributed Large Language Models (2024.findings-acl)

Copied to clipboard

Challenge: despite impressive performance, large language models still struggle with hallucinations . current approaches suffer from suboptimal citation quality due to reliance on in-context learning .
Approach: They propose a framework that teaches large language models to generate fine-grained citations.
Outcome: The proposed framework outperforms all baselines on the ALCE benchmark and achieves an average improvement of 14.21% in citation quality.
A.S.E: A Repository-Level Benchmark for Evaluating Security in AI-Generated Code (2026.findings-acl)

Copied to clipboard

Challenge: Existing security evaluation benchmarks lack relevance to real-world AI programming tasks . current LLMs struggle with secure coding, research shows .
Approach: They propose a repository-level evaluation benchmark to assess security of AI-generated code.
Outcome: The proposed framework mirrors real-world AI programming tasks and offers valuable insights into the state of AI code generation.
Improving Contextual Faithfulness of Large Language Models via Retrieval Heads-Induced Optimization (2025.acl-long)

Copied to clipboard

Challenge: Existing frameworks for retrieval-augmented large language models (LLMs) are lacking in LFQA faithfulness testing.
Approach: They propose a framework to teach retrieval-augmented large language models to explicitly discriminate between faithful and unfaithful generations.
Outcome: The proposed framework outperforms GPT-4o in LFQA scenarios and outperformed existing benchmarks.

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