papers

Publications (20)

cs.CL2026

ATTS: Asynchronous Test-Time Scaling via Conformal Prediction

Jing Xiong, Qiujiang Chen, Fanghua Ye +11

Large language models (LLMs) benefit from test-time scaling but are often hampered by high inference latency. Speculative decoding is a natural way to accelerate the scaling proces…

cs.CL2023

VCSUM: A Versatile Chinese Meeting Summarization Dataset

Han Wu, Mingjie Zhan, Haochen Tan +3

Compared to news and chat summarization, the development of meeting summarization is hugely decelerated by the limited data. To this end, we introduce a versatile Chinese meeting s…

cs.CL2023

Learning Locality and Isotropy in Dialogue Modeling

Han Wu, Haochen Tan, Mingjie Zhan +4

Existing dialogue modeling methods have achieved promising performance on various dialogue tasks with the aid of Transformer and the large-scale pre-trained language models. Howeve…

cs.CL2025

MHPP: Exploring the Capabilities and Limitations of Language Models Beyond Basic Code Generation

Jianbo Dai, Jianqiao Lu, Yunlong Feng +6

Recent advancements in large language models (LLMs) have greatly improved code generation, specifically at the function level. For instance, GPT-4o has achieved a 91.0\% pass rate…

cs.CL2026

OVD: On-policy Verbal Distillation

Jing Xiong, Hui Shen, Shansan Gong +7

Knowledge distillation offers a promising path to transfer reasoning capabilities from large teacher models to efficient student models; however, existing token-level on-policy dis…

cs.CL2025

DeepDiver: Adaptive Search Intensity Scaling via Open-Web Reinforcement Learning

Wenxuan Shi, Haochen Tan, Chuqiao Kuang +7

Information seeking demands iterative evidence gathering and reflective reasoning, yet large language models (LLMs) still struggle with it in open-web question answering. Existing…

cs.CL2025

CiteCheck: Towards Accurate Citation Faithfulness Detection

Ziyao Xu, Shaohang Wei, Zhuoheng Han +6

Citation faithfulness detection is critical for enhancing retrieval-augmented generation (RAG) systems, yet large-scale Chinese datasets for this task are scarce. Existing methods…

cs.SD2026

DSA-Tokenizer: Disentangled Semantic-Acoustic Tokenization via Flow Matching-based Hierarchical Fusion

Hanlin Zhang, Daxin Tan, Dehua Tao +5

Speech tokenizers are a key building block of fully discrete Speech LLMs. Existing tokenizers either prioritize semantic encoding, fuse semantic content with acoustic style insepar…

cs.CL2022

Zero-shot Cross-lingual Conversational Semantic Role Labeling

Han Wu, Haochen Tan, Kun Xu +3

While conversational semantic role labeling (CSRL) has shown its usefulness on Chinese conversational tasks, it is still under-explored in non-Chinese languages due to the lack of…

cs.AI2025

TIME: A Multi-level Benchmark for Temporal Reasoning of LLMs in Real-World Scenarios

Shaohang Wei, Wei Li, Feifan Song +5

Temporal reasoning is pivotal for Large Language Models (LLMs) to comprehend the real world. However, existing works neglect the real-world challenges for temporal reasoning: (1) i…

cs.LG2026

GRPO-VPS: Enhancing Group Relative Policy Optimization with Verifiable Process Supervision for Effective Reasoning

Jingyi Wang, Lei Zhu, Tengjin Weng +8

Reinforcement Learning with Verifiable Rewards (RLVR) has advanced the reasoning capabilities of Large Language Models (LLMs) by leveraging direct outcome verification instead of l…

cs.CL2024

MR-Ben: A Meta-Reasoning Benchmark for Evaluating System-2 Thinking in LLMs

Zhongshen Zeng, Yinhong Liu, Yingjia Wan +16

Large language models (LLMs) have shown increasing capability in problem-solving and decision-making, largely based on the step-by-step chain-of-thought reasoning processes. Howeve…

cs.CL2022

A Sentence is Worth 128 Pseudo Tokens: A Semantic-Aware Contrastive Learning Framework for Sentence Embeddings

Haochen Tan, Wei Shao, Han Wu +2

Contrastive learning has shown great potential in unsupervised sentence embedding tasks, e.g., SimCSE. However, We find that these existing solutions are heavily affected by superf…

cs.CL2020

Semantic Role Labeling Guided Multi-turn Dialogue ReWriter

Kun Xu, Haochen Tan, Linfeng Song +4

For multi-turn dialogue rewriting, the capacity of effectively modeling the linguistic knowledge in dialog context and getting rid of the noises is essential to improve its perform…

cs.CL2024

Towards Versatile and Efficient Visual Knowledge Integration into Pre-trained Language Models with Cross-Modal Adapters

Xinyun Zhang, Haochen Tan, Han Wu +1

Humans learn language via multi-modal knowledge. However, due to the text-only pre-training scheme, most existing pre-trained language models (PLMs) are hindered from the multi-mod…

cs.CL2023

Reconstruct Before Summarize: An Efficient Two-Step Framework for Condensing and Summarizing Meeting Transcripts

Haochen Tan, Han Wu, Wei Shao +5

Meetings typically involve multiple participants and lengthy conversations, resulting in redundant and trivial content. To overcome these challenges, we propose a two-step framewor…

cs.CL2024

PROXYQA: An Alternative Framework for Evaluating Long-Form Text Generation with Large Language Models

Haochen Tan, Zhijiang Guo, Zhan Shi +8

Large Language Models (LLMs) have succeeded remarkably in understanding long-form contents. However, exploring their capability for generating long-form contents, such as reports a…

eess.AS2026

SpeechEditBench: A Bilingual Multi-Attribute Benchmark for Instruction-Guided Speech Editing

Hanlin Zhang, Daxin Tan, Dehua Tao +3

Instruction-guided speech editing requires a model to modify specified speech attributes while preserving unrelated characteristics. Despite rapid progress in Speech Large Language…

cs.LG2026

Merging Beyond: Streaming LLM Updates via Activation-Guided Rotations

Yuxuan Yao, Haonan Sheng, Qingsong Lv +11

The escalating scale of Large Language Models (LLMs) necessitates efficient adaptation techniques. Model merging has gained prominence for its efficiency and controllability. Howev…

cs.CL2025

FaStfact: Faster, Stronger Long-Form Factuality Evaluations in LLMs

Yingjia Wan, Haochen Tan, Xiao Zhu +9

Evaluating the factuality of long-form generations from Large Language Models (LLMs) remains challenging due to efficiency bottlenecks and reliability concerns. Prior efforts attem…