Papers by Zhiwei Xu

19 papers
Reinforced Efficient Reasoning via Semantically Diverse Exploration (2026.acl-long)

Copied to clipboard

Challenge: Existing methods for reinforcement learning with verifiable rewards suffer from limited exploration diversity and inefficient reasoning.
Approach: They propose a method that rewards concise and correct reasoning while penalizing unnecessarily long reasoning chains.
Outcome: Extensive experiments on Qwen and Llama models validate the effectiveness and efficiency of ROSE.
Draft Model Knows When to Stop: Self-Verification Speculative Decoding for Long-Form Generation (2025.emnlp-main)

Copied to clipboard

Challenge: Conventional speculative decoding methods use a predefined length policy for proposing drafts, but the reality deviates from this assumption.
Approach: They propose a self-verification length policy that adaptively determines the lengths of draft sequences by referring to the draft entropy.
Outcome: The proposed method achieves 17% speedup on MT-Bench and 22% speedup in long-form reasoning.
RevMUX: Data Multiplexing with Reversible Adapters for Efficient LLM Batch Inference (2024.emnlp-main)

Copied to clipboard

Challenge: Large language models (LLMs) have brought a great breakthrough to the natural language processing community, but their high throughput demands make them difficult to handle concurrent queries.
Approach: They propose a parameter-efficient data multiplexing framework that integrates a reversible design in the multiplexer and can be reused to perform reverse operations and restore individual samples for classification.
Outcome: The proposed framework improves inference efficiency while maintaining satisfactory classification performance.
Bridging the Capability Gap: Joint Alignment Tuning for Harmonizing LLM-based Multi-Agent Systems (2025.findings-emnlp)

Copied to clipboard

Challenge: Existing methods for fine-tuning agents are often inadequate . a multi-agent system can solve complex tasks by dividing responsibilities among specialized agents .
Approach: a new framework is proposed to improve agents collaboration through iterative alignment.
Outcome: The proposed framework outperforms state-of-the-art benchmarks on held-in and held-out tasks.
PairCFR: Enhancing Model Training on Paired Counterfactually Augmented Data through Contrastive Learning (2024.acl-long)

Copied to clipboard

Challenge: Recent research shows that training with CAD may lead models to overly focus on modified features while ignoring other important contextual information.
Approach: They propose to use contrastive learning to promote global feature alignment and learning counterfactual clues to improve model performance.
Outcome: The proposed method outperforms the state-of-the-art on out-of distribution (OOD) datasets.
Same Claim, Different Judgment: Benchmarking Scenario-Induced Bias in Multilingual Financial Misinformation Detection (2026.findings-acl)

Copied to clipboard

Challenge: Existing research on LLM biases has focused on direct questioning or general-purpose settings . pronounced behavioral biase despite their growing deployment in financial analysis, forecasting, and decision support.
Approach: They propose a benchmark to evaluate behavioral biases of large language models in MFMD . they use a multilingual financial misinformation dataset to integrate these with misinformation claims .
Outcome: The proposed benchmark evaluates behavioral biases of large language models across economic scenarios.
Efficient Cross-Task Prompt Tuning for Few-Shot Conversational Emotion Recognition (2023.findings-emnlp)

Copied to clipboard

Challenge: Emotion Recognition in Conversation (ERC) models are often expensive to train and fine-tune .
Approach: They propose a derivative-free optimization method for few-shot conversational emotion recognition that leverages sharable cross-task knowledge by exploiting external knowledge from other source tasks.
Outcome: The proposed method improves on few-shot scenarios and zero-shot transfers on five different contextual conversation datasets.
Can Large Language Models Mine Interpretable Financial Factors More Effectively? A Neural-Symbolic Factor Mining Agent Model (2024.findings-acl)

Copied to clipboard

Challenge: Existing factor mining models are inefficient and inefficient, resulting in a significant challenge to extract interpretable factors.
Approach: They propose a model that integrates the strengths of both neural and symbolic models for factor mining.
Outcome: The proposed model surpasses the SOTA RankIC and RankICIR in predicting S&P 500 returns on real-world stock market data.
Understanding the Language Model to Solve the Symbolic Multi-Step Reasoning Problem from the Perspective of Buffer Mechanism (2025.findings-emnlp)

Copied to clipboard

Challenge: Large language models struggle with complex reasoning tasks, such as mathematical problem-solving.
Approach: They constructed a symbolic multi-step reasoning task to investigate the information propagation mechanisms in Transformer models when solving the task through direct answering and Chain-of-Thought (CoT) reasoning.
Outcome: The proposed algorithm improves on 7 multi-step reasoning datasets, while introducing only 132 trainable parameters.
All That Glisters Is Not Gold: A Benchmark for Reference-Free Counterfactual Financial Misinformation Detection (2026.acl-long)

Copied to clipboard

Challenge: RFC-Bench evaluates large language models on financial misinformation under realistic news . current models struggle to maintain coherent belief states without external grounding, study finds .
Approach: They propose a benchmark for evaluating large language models on financial misinformation under realistic news.
Outcome: The proposed model performs better when context is available, while reference-free settings expose significant weaknesses.
FaStFact: Faster, Stronger Long-Form Factuality Evaluations in LLMs (2025.findings-emnlp)

Copied to clipboard

Challenge: Prior evaluation pipelines fail to evaluate factuality of long-form LLMs due to inefficiency and costly human assessment.
Approach: They propose a fast and strong evaluation pipeline that can evaluate factuality of long-form LLMs . they propose 'faStFact' to reduce cost of web searching and inference calling .
Outcome: The proposed evaluation pipeline achieves highest alignment with human evaluation and efficiency among existing baselines.
Unraveling the Mystery: Defending Against Jailbreak Attacks Via Unearthing Real Intention (2025.coling-main)

Copied to clipboard

Challenge: Large Language Models (LLMs) are increasingly vulnerable to elusive and implicit intentions, causing security risks and compromising user experience.
Approach: They propose a method to detect and mitigate implicit jailbreak attacks using LLMs by unearthing real intentions and a greedy gradient-based algorithm to remove the least important parts of a sentence.
Outcome: The proposed method reduces attacks success rate and Harmful Score while maintaining overall model performance.
R-Judge: Benchmarking Safety Risk Awareness for LLM Agents (2024.findings-emnlp)

Copied to clipboard

Challenge: Large language models (LLMs) have shown compelling abilities in reasoning, decision-making, and instruction following.
Approach: They propose a benchmark to evaluate the proficiency of large language models (LLMs) in judging and identifying safety risks given agent interaction records.
Outcome: The proposed model outperforms the best-performing model, GPT-4o, while no other models significantly exceed the random.
Unsupervised Sign Language Translation and Generation (2024.findings-acl)

Copied to clipboard

Challenge: Experimental results on the BBC-Oxford Sign Language dataset reveal that USLNet achieves competitive results compared to supervised baseline models.
Approach: They propose an unsupervised sign language translation and generation network that learns from abundant single-modality data without parallel sign language data.
Outcome: The proposed model achieves competitive results compared to baseline models on the BBC-Oxford Sign Language dataset and Open-Domain American Sign Language data.
SLIM: Subtrajectory-Level Elimination for More Effective Reasoning (2025.findings-emnlp)

Copied to clipboard

Challenge: Notable examples include OpenAI’s o1/o3/o4 series and DeepSeek-R1 .
Approach: They develop a framework to identify suboptimal subtrajectories based on human-established criteria . they also use a sampling algorithm to select data whose reasoning process is free from suboptimally subtravertories to the highest degree .
Outcome: The proposed method reduces the number of suboptimal subtrajectories by 25.9% during the inference process.
SoftCoT: Soft Chain-of-Thought for Efficient Reasoning with LLMs (2025.acl-long)

Copied to clipboard

Challenge: Existing approaches to continuous-space reasoning focus on hard token decoding and suffer from catastrophic forgetting.
Approach: They propose a method that generates instance-specific soft thought tokens as the initial chain of thoughts and maps them into the LLM’s representation space via a trainable projection module.
Outcome: The proposed method improves LLM reasoning performance through supervised, parameter-efficient fine-tuning.
xLAM: A Family of Large Action Models to Empower AI Agent Systems (2025.naacl-long)

Copied to clipboard

Challenge: Autonomous agents powered by large language models (LLMs) have attracted significant research interest, but there are few standards for developing specialized models for agent tasks.
Approach: They propose a series of large action models with dense and mixture-of-expert architectures that unifies, augments, and synthesizes diverse datasets to enhance agent generalizability and performance.
Outcome: The proposed models outperform GPT-4, Claude-3, and many other models in terms of tool use and outperformed GPT-based models on multiple agent ability benchmarks.
A Survey on Natural Language Counterfactual Generation (2024.findings-emnlp)

Copied to clipboard

Challenge: Recent advances in NLP are driven by a variety of Large Language Models (LLMs), such as GPT-3 (175B) and PaLM (540B).
Approach: They propose a taxonomy that categorizes the methods into four groups and summarizes the metrics for evaluating the generation quality.
Outcome: The proposed taxonomy categorizes the generation methods into four groups and summarizes the metrics for evaluating the quality.
CONSTRUCTURE: Benchmarking CONcept STRUCTUre REasoning for Multimodal Large Language Models (2024.findings-emnlp)

Copied to clipboard

Challenge: Existing multimodal large language models lack the ability to perceive the visual world with a deep concept structure cognition.
Approach: They propose a concept-level benchmark to assess MLLMs’ hierarchical concept understanding and reasoning abilities.
Outcome: The proposed model outperforms state-of-the-art models in concept structure reasoning evaluation.

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