Papers by Wen Tao

25 papers
There Are a Thousand Hamlets in a Thousand People’s Eyes: Enhancing Knowledge-grounded Dialogue with Personal Memory (2022.acl-long)

Copied to clipboard

Challenge: Existing methods for knowledge selection focus on relevance between knowledge and dialogue context, ignoring personal preference for knowledge.
Approach: They propose to introduce personal memory into knowledge selection in chatbots to address personalization issue by integrating personal memory and inverse mapping into a closed loop.
Outcome: The proposed method outperforms existing methods significantly on automatic evaluation and human evaluation.
A Pre-training Strategy for Zero-Resource Response Selection in Knowledge-Grounded Conversations (2021.acl-long)

Copied to clipboard

Challenge: Existing methods to train retrieval-based dialogue systems rely on crowd-sourced data . however, it is difficult to collect large-scale dialogues that are grounded on background knowledge .
Approach: They propose to decompose training of knowledge-grounded response selection into three tasks . they propose to combine query-passage matching task with query-dialogue history matching task .
Outcome: Experimental results show that the proposed model can perform comparable to existing methods . the retrieval-based system can leverage background knowledge when conversing with humans .
Two-Stage Regularization-Based Structured Pruning for LLMs (2026.acl-long)

Copied to clipboard

Challenge: Structural pruning is a promising solution for large language models . prior structured pruning methods remove unimportant parameters based on certain metrics .
Approach: They propose a structural pruning method that iteratively learns the weights of transformer layers by adding their l1-norm to the loss function.
Outcome: The proposed pruning method outperforms strong layer-wise pruning methods without requiring retraining.
ReFL: Reflective Feedback Learning for Hallucination Detection of Large Language Models (2026.acl-long)

Copied to clipboard

Challenge: Existing methods for detecting hallucinations depend on external knowledge sources, incurring high computational costs and limiting real-time applicability, or extract the model’s internal states, leading to poor generalization.
Approach: They propose a hallucination detection framework that leverages corrective in-context learning to guide LLMs to recognize their own prediction errors and adjust internal representations, critically without updating model weights.
Outcome: The proposed framework outperforms existing methods on two benchmark datasets and achieves state-of-the-art performance.
Listen, Watch, and Learn to Feel: Retrieval-Augmented Emotion Reasoning for Compound Emotion Generation (2025.findings-acl)

Copied to clipboard

Challenge: Existing methods to assess human emotion are limited by the subjective nature of emotion perception, limiting the robustness of existing models.
Approach: They propose a plug-and-play module that enhances MLLMs’ ability to tackle compound and context-rich emotion tasks.
Outcome: The proposed framework improves MLLMs' ability to tackle compound and context-rich emotion tasks and the Compound Emotion QA dataset shows it performs well across both benchmarks and evaluation frameworks.
Multi-Programming Language Sandbox for LLMs (2025.acl-demo)

Copied to clipboard

Challenge: MPLSandbox is an out-of-the-box multi-programming language sandbox designed to provide unified and comprehensive feedback from compiler and analysis tools for Large Language Models (LLMs).
Approach: They propose a multi-programming language sandbox that provides unified feedback from compilers and analysis tools for Large Language Models.
Outcome: The proposed multi-language sandbox can provide comprehensive feedback from compilers and analysis tools for large language models (LLMs).
Enhancing Online Recruitment with Category-Aware MoE and LLM-based Data Augmentation (2026.acl-industry)

Copied to clipboard

Challenge: Existing methods to measure the matching degree of a job and a candidate face several challenges, such as low-quality job descriptions and similar candidate-job pairs.
Approach: They propose a large language model-based method that polishes and rewrites low-quality job descriptions by leveraging chain-of-thought prompts and category-aware Mixture of Experts (MoE) module incorporates category embeddings to dynamically assign weights to the experts and learns more distinguishable patterns for similar candidate-job pairs.
Outcome: The proposed method surpasses existing methods by 2.40% in AUC and 7.46% in GAUC and boosts click-through conversion rate (CTCVR) by 19.4% in online tests, saving millions of CNY in external headhunting expenses.
SPARK: Strategic Policy-Aware Exploration via Dynamic Branching for Long-Horizon Agentic Learning (2026.acl-long)

Copied to clipboard

Challenge: Existing methods for training large language models waste computation budget on trivial steps while failing to guarantee sample quality.
Approach: They propose a framework that selectively branches at critical decision states for resource-efficient exploration.
Outcome: The proposed framework activates adaptive branching exploration at critical decision states to probe promising trajectories, thereby achieving precise resource allocation that prioritizes sampling quality over blind coverage.
PACE: Prefix-Protected and Difficulty-Aware Compression for Efficient Reasoning (2026.findings-acl)

Copied to clipboard

Challenge: Existing LRMs often suffer from "overthinking" and excessively long reasoning traces . a dual-level framework for length compression of LRM is proposed .
Approach: They propose a framework for prefix-protected and difficulty-aware compression under hierarchical supervision.
Outcome: The proposed framework reduces token usage while improving accuracy on math benchmarks.
MARS-Bench: A Multi-turn Athletic Real-world Scenario Benchmark for Dialogue Evaluation (2025.findings-emnlp)

Copied to clipboard

Challenge: Large Language Models (LLMs) have been widely adopted in real-world dialogue applications, but their robustness is criticized all along.
Approach: They propose to use play-by-play text commentary to build a multi-turn athletic real-world scenario dialogue benchmark to evaluate three critical aspects of multi-turned conversations: ultra multi- turn, interactive multi-twist, and cross-turn tasks.
Outcome: The proposed benchmarks outperform open-source LLMs on three critical aspects of multi-turn conversations: ultra multi-turned, interactive multi- turn, and cross-turn tasks.
SpecCache: Speculative KV Cache Reuse for Efficient RAG Serving (2026.acl-long)

Copied to clipboard

Challenge: Retrieval-Augmented Generation (RAG) improves LLMs but faces high prefill latency during long contexts.
Approach: They propose a method that uses deep-layer hidden-state norms to guide token selection . they propose to use deep-layered hidden-status norms as a proxy to guide the token selection.
Outcome: The proposed SpecCache outperforms state-of-the-art (SOTA) benchmarks.
Unlocking Data-free Low-bit Quantization with Matrix Decomposition for KV Cache Compression (2024.acl-long)

Copied to clipboard

Challenge: Existing methods to compress KV cache compromise precision or require extra data for calibration, limiting their practicality in LLM deployment.
Approach: They propose a low-bit quantization technique based on tensor decomposition to effectively compress KV cache.
Outcome: The proposed method reduces memory footprint and performance by 75% . it is compared with existing methods that compromise precision or require extra data for calibration .
RadialRouter: Structured Representation for Efficient and Robust Large Language Models Routing (2025.findings-emnlp)

Copied to clipboard

Challenge: Current routing methods are limited in exploring the connection between query and LLM characteristics.
Approach: They propose a framework for LLM routing that uses a transformer-based backbone and a radial structure to articulate the query-LLMs relationship.
Outcome: The proposed framework outperforms existing routing methods by 9.2% and 5.8% on RouterBench.
d-TreeRPO: Towards More Reliable Policy Optimization for Diffusion Language Models (2026.acl-long)

Copied to clipboard

Challenge: Existing RL methods suffer from reliability bottlenecks due to reward sparsity and intractable computations . d-TreeRPO provides fine-grained and verifiable step-wise reward signals .
Approach: They propose a reliable reinforcement learning framework for diffusion large language models that leverages tree-structured rollouts and bottom-up advantage computation based on verifiable outcome rewards.
Outcome: The proposed framework outperforms baseline models and achieves significant improvements across reasoning benchmarks.
TemplateRL: Structured Template-Guided Reinforcement Learning for LLM Reasoning (2026.findings-acl)

Copied to clipboard

Challenge: Existing RL methods rely on unstructured self-sampling to fit scalar rewards, resulting in inefficient rollouts.
Approach: They propose a structured template-guided RL framework that augments policy optimization with explicit template guidance.
Outcome: Experiments show that TemplateRL outperforms GRPO and GRPI by 99% on AIME and 41% on AMC with superior stability on weak models and remarkable cross-domain generalization.
Learn to Memorize: Scalable Continual Learning in Semiparametric Models with Mixture-of-Neighbors Induction Memory (2025.acl-long)

Copied to clipboard

Challenge: Semiparametric language models (LMs) use static storage, which lacks learning capability and is disconnected from the internal information flow of the parametric models.
Approach: They reconceptualize the non-parametric memory represented by kNN-LM as a learnable Mixture-of-Neighbors Induction Memory (MoNIM) this synergizes the induction capabilities of attention heads with the memorization strength of feed-forward networks .
Outcome: The proposed model is a learnable Mixture-of-neighbors induction memory (MoNIM) it synergizes the induction capabilities of attention heads with the memorization strength of feed-forward networks (FFNs).
Beyond Examples: Towards Automated Thought-level In-Context Reasoning for Large Language Models (2026.acl-long)

Copied to clipboard

Challenge: In-context learning (ICL) struggles with complex reasoning due to superficial, example-level implicit imitation.
Approach: They propose an automated method that shifts from surface-level examples to more guidance-oriented thought patterns.
Outcome: The proposed method achieves 80.6% accuracy on MATH and 62.5% on AMC, surpassing GPT-4o’s 77.2% and 57.5% accuracy.
How to Make Large Language Models Generate 100% Valid Molecules? (2025.emnlp-main)

Copied to clipboard

Challenge: Large language models (LLMs) can learn to perform a wide range of tasks, but generating valid molecules using representations like SMILES is challenging in few-shot settings.
Approach: They propose a language framework that converts invalid SMILES to SELFIES and LLMs as post-hoc correctors to ensure that the molecules generated by LLM are 100% valid.
Outcome: The proposed model performs worse with SELFIES than with SMILES and improves on other metrics.
Rethinking Table Pruning in TableQA: From Sequential Revisions to Gold Trajectory-Supervised Parallel Search (2026.acl-long)

Copied to clipboard

Challenge: Existing pruning methods rely on sequential revisions and unreliable critique signals . Existing methods fail to detect the loss of answer-critical data .
Approach: They propose a table pruning framework which transforms table pruning to gold trajectory-supervised parallel search.
Outcome: The proposed framework outperforms the strongest baseline pruning framework by 3.2% on various tabular reasoning tasks.
Domain Adaptation for Person-Job Fit with Transferable Deep Global Match Network (D19-1)

Copied to clipboard

Challenge: Existing methods to solve person-job fit in single-domain setting are limited by labeled data.
Approach: They propose a deep global match network for capturing the global semantic interactions between two sentences from a job posting and a candidate resume respectively.
Outcome: The proposed model is effective when there is not enough labeled data.
GLiM: Integrating Graph Transformer and LLM for Document-Level Biomedical Relation Extraction with Incomplete Labeling (2025.findings-acl)

Copied to clipboard

Challenge: Document-level relation extraction (DocRE) solves problems of document quality . number of entities and entity-pair relations increases, causing incomplete annotations .
Approach: a framework that reduces the problem space using a graph-enhanced Transformer-based model is proposed . GLiM leverages large language models for reasoning to reduce the problem-space .
Outcome: GLiM boosts average recall and F1 scores on biomedical datasets . compared with existing models, GLim outperforms existing models on biomedicine benchmarks compared to existing models .
ATLAS: Orchestrating Heterogeneous Models and Tools for Multi-Domain Complex Reasoning (2026.findings-acl)

Copied to clipboard

Challenge: Existing approaches to optimize large language models with external tools are limited.
Approach: They propose a dual-path framework for dynamic tool usage in cross-domain complex reasoning . they exploit empirical priors for domain-specific alignment and RL-based multi-step routing .
Outcome: The proposed framework outperforms closed-source models and existing methods on in-distribution and out-of-distortion tasks.
CAIR: Causal Adaptive Information-based Reinforcement Learning for Multimodal Emotion Reasoning (2026.findings-acl)

Copied to clipboard

Challenge: Existing methods for multimodal emotion reasoning produce fluent but superficial explanations that lack authentic logical derivation.
Approach: They propose a framework that treats rationales as causal mediators between raw perceptual signals and emotional semantics and an adaptive optimization mechanism to balance perception and reasoning across varying cognitive loads.
Outcome: The proposed framework outperforms specialized SFT models by 14.4% while enhancing rationale faithfulness.
When Agents Look the Same: Quantifying Distillation-Induced Similarity in Tool-Use Behaviors (2026.acl-long)

Copied to clipboard

Challenge: Existing metrics fail to distinguish mandatory behaviors required for task success from non-mandatory patterns that reflect a model’s autonomous preferences.
Approach: They propose to use response pattern similarity and action graph similarity to isolate non-mandatory behaviors from mandatory behaviors.
Outcome: Evaluating 18 models from 8 providers on -Bench and 2-Bench against Claude Sonnet 4.5, the authors find that within-family model pairs score 5.9 pp higher in response pattern similarity and action graph similarity .
Scaling Law for Multimodal Large Language Model Supervised Fine-Tuning (2026.acl-long)

Copied to clipboard

Challenge: supervised fine-tuning (SFT) is crucial for multimodal large language models, yet a comprehensive scaling law is lacking . et al.: scaling laws focus on model size, pre-training tokens, and MLLM SFT data volumes .
Approach: They propose two scaling laws to guide optimal model-data configuration . they propose one applicable when training data volumes are well defined by researchers .
Outcome: The proposed scaling laws provide valuable recommendations for optimal resource allocation . they show that the proposed laws are more accurate than existing models .

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