Publications (69)
DolphCoder: Echo-Locating Code Large Language Models with Diverse and Multi-Objective Instruction Tuning
Yejie Wang, Keqing He, Guanting Dong +8
Code Large Language Models (Code LLMs) have demonstrated outstanding performance in code-related tasks. Several instruction tuning approaches have been proposed to boost the code g…
SmartSearch: Process Reward-Guided Query Refinement for Search Agents
Tongyu Wen, Guanting Dong, Zhicheng Dou
Large language model (LLM)-based search agents have proven promising for addressing knowledge-intensive problems by incorporating information retrieval capabilities. Existing works…
Towards On-Policy Data Evolution for Visual-Native Multimodal Deep Search Agents
Shijue Huang, Hangyu Guo, Guanting Dong +8
Multimodal deep search requires an agent to solve open-world problems by chaining search, tool use, and visual reasoning over evolving textual and visual context. Two bottlenecks l…
WebThinker: Empowering Large Reasoning Models with Deep Research Capability
Xiaoxi Li, Jiajie Jin, Guanting Dong +5
Large reasoning models (LRMs), such as OpenAI-o1 and DeepSeek-R1, demonstrate impressive long-horizon reasoning capabilities. However, their reliance on static internal knowledge l…
Agent-World: Scaling Real-World Environment Synthesis for Evolving General Agent Intelligence
Guanting Dong, Junting Lu, Junjie Huang +17
Large language models are increasingly expected to serve as general-purpose agents that interact with external, stateful tool environments. The Model Context Protocol (MCP) and bro…
Qwen2 Technical Report
An Yang, Baosong Yang, Binyuan Hui +59
This report introduces the Qwen2 series, the latest addition to our large language models and large multimodal models. We release a comprehensive suite of foundational and instruct…
Agentic Entropy-Balanced Policy Optimization
Guanting Dong, Licheng Bao, Zhongyuan Wang +11
Recently, Agentic Reinforcement Learning (Agentic RL) has made significant progress in incentivizing the multi-turn, long-horizon tool-use capabilities of web agents. While mainstr…
Hierarchical Document Refinement for Long-context Retrieval-augmented Generation
Jiajie Jin, Xiaoxi Li, Guanting Dong +6
Real-world RAG applications often encounter long-context input scenarios, where redundant information and noise results in higher inference costs and reduced performance. To addres…
HiRA: A Hierarchical Reasoning Framework for Decoupled Planning and Execution in Deep Search
Jiajie Jin, Xiaoxi Li, Guanting Dong +5
Complex information needs in real-world search scenarios demand deep reasoning and knowledge synthesis across diverse sources, which traditional retrieval-augmented generation (RAG…
ET-Agent: Incentivizing Effective Tool-Integrated Reasoning Agent via Behavior Calibration
Yifei Chen, Guanting Dong, Zhicheng Dou
Large Language Models (LLMs) can extend their parameter knowledge limits by adopting the Tool-Integrated Reasoning (TIR) paradigm. However, existing LLM-based agent training framew…
FlashRAG: A Modular Toolkit for Efficient Retrieval-Augmented Generation Research
Jiajie Jin, Yutao Zhu, Guanting Dong +7
With the advent of large language models (LLMs) and multimodal large language models (MLLMs), the potential of retrieval-augmented generation (RAG) has attracted considerable resea…
Self-play with Execution Feedback: Improving Instruction-following Capabilities of Large Language Models
Guanting Dong, Keming Lu, Chengpeng Li +4
One core capability of large language models (LLMs) is to follow natural language instructions. However, the issue of automatically constructing high-quality training data to enhan…
EnvScaler: Scaling Tool-Interactive Environments for LLM Agent via Programmatic Synthesis
Xiaoshuai Song, Haofei Chang, Guanting Dong +3
Large language models (LLMs) are expected to be trained to act as agents in various real-world environments, but this process relies on rich and varied tool-interaction sandboxes.…
PreAct: Prediction Enhances Agent's Planning Ability
Dayuan Fu, Jianzhao Huang, Siyuan Lu +4
Addressing the disparity between forecasts and actual results can enable individuals to expand their thought processes and stimulate self-reflection, thus promoting accurate planni…
Multi-Perspective Consistency Enhances Confidence Estimation in Large Language Models
Pei Wang, Yejie Wang, Muxi Diao +3
In the deployment of large language models (LLMs), accurate confidence estimation is critical for assessing the credibility of model predictions. However, existing methods often fa…
Semi-Supervised Knowledge-Grounded Pre-training for Task-Oriented Dialog Systems
Weihao Zeng, Keqing He, Zechen Wang +8
Recent advances in neural approaches greatly improve task-oriented dialogue (TOD) systems which assist users to accomplish their goals. However, such systems rely on costly manuall…
Towards Robust and Generalizable Training: An Empirical Study of Noisy Slot Filling for Input Perturbations
Jiachi Liu, Liwen Wang, Guanting Dong +8
In real dialogue scenarios, as there are unknown input noises in the utterances, existing supervised slot filling models often perform poorly in practical applications. Even though…
How Abilities in Large Language Models are Affected by Supervised Fine-tuning Data Composition
Guanting Dong, Hongyi Yuan, Keming Lu +7
Large language models (LLMs) with enormous pre-training tokens and parameters emerge diverse abilities, including math reasoning, code generation, and instruction following. These…
Revisit Out-Of-Vocabulary Problem for Slot Filling: A Unified Contrastive Frameword with Multi-level Data Augmentations
Daichi Guo, Guanting Dong, Dayuan Fu +9
In real dialogue scenarios, the existing slot filling model, which tends to memorize entity patterns, has a significantly reduced generalization facing Out-of-Vocabulary (OOV) prob…
Towards Verifiable Multimodal Deep Research: A Multi-Agent Harness for Interleaved Report Generation
Chenghao Zhang, Guanting Dong, Yufan Liu +3
Large Language Models (LLMs) have advanced autonomous agents from deep search, which retrieves concise factual answers, to deep research, which synthesizes scattered evidence into…
DeepAgent: A General Reasoning Agent with Scalable Toolsets
Xiaoxi Li, Wenxiang Jiao, Jiarui Jin +8
Large reasoning models have demonstrated strong problem-solving abilities, yet real-world tasks often require external tools and long-horizon interactions. Existing agent framework…
WebSwarm: Recursive Multi-Agent Orchestration for Deep-and-Wide Web Search
Xiaoshuai Song, Liancheng Zhang, Kangzhi Zhao +8
Large language model (LLM)-based web search agents are transforming information seeking from simple factoid question answering into complex, deep-and-wide search and research-orien…
V-Thinker: Interactive Thinking with Images
Runqi Qiao, Qiuna Tan, Minghan Yang +11
Empowering Large Multimodal Models (LMMs) to deeply integrate image interaction with long-horizon reasoning capabilities remains a long-standing challenge in this field. Recent adv…
Agentic Reinforced Policy Optimization
Guanting Dong, Hangyu Mao, Kai Ma +11
Large-scale reinforcement learning with verifiable rewards (RLVR) has demonstrated its effectiveness in harnessing the potential of large language models (LLMs) for single-turn rea…
OccuQuest: Mitigating Occupational Bias for Inclusive Large Language Models
Mingfeng Xue, Dayiheng Liu, Kexin Yang +5
The emergence of large language models (LLMs) has revolutionized natural language processing tasks. However, existing instruction-tuning datasets suffer from occupational bias: the…
Noise-BERT: A Unified Perturbation-Robust Framework with Noise Alignment Pre-training for Noisy Slot Filling Task
Jinxu Zhao, Guanting Dong, Yueyan Qiu +4
In a realistic dialogue system, the input information from users is often subject to various types of input perturbations, which affects the slot-filling task. Although rule-based…
MuggleMath: Assessing the Impact of Query and Response Augmentation on Math Reasoning
Chengpeng Li, Zheng Yuan, Hongyi Yuan +6
In math reasoning with large language models (LLMs), fine-tuning data augmentation by query evolution and diverse reasoning paths is empirically verified effective, profoundly narr…
Multi-Dimensional Insights: Benchmarking Real-World Personalization in Large Multimodal Models
YiFan Zhang, Shanglin Lei, Runqi Qiao +10
The rapidly developing field of large multimodal models (LMMs) has led to the emergence of diverse models with remarkable capabilities. However, existing benchmarks fail to compreh…
DecIF: Improving Instruction-Following through Meta-Decomposition
Tingfeng Hui, Pengyu Zhu, Bowen Ping +4
Instruction-following has emerged as a crucial capability for large language models (LLMs). However, existing approaches often rely on pre-existing documents or external resources…
How Do Your Code LLMs Perform? Empowering Code Instruction Tuning with High-Quality Data
Yejie Wang, Keqing He, Dayuan Fu +11
Recently, there has been a growing interest in studying how to construct better code instruction tuning data. However, we observe Code models trained with these datasets exhibit hi…
Large Language Models Meet Open-World Intent Discovery and Recognition: An Evaluation of ChatGPT
Xiaoshuai Song, Keqing He, Pei Wang +6
The tasks of out-of-domain (OOD) intent discovery and generalized intent discovery (GID) aim to extend a closed intent classifier to open-world intent sets, which is crucial to tas…
Toward Generalist Autonomous Research via Hypothesis-Tree Refinement
Jiajie Jin, Yuyang Hu, Kai Qiu +15
Scientific progress depends on a repeated loop of exploration, experimentation, and abstraction. Researchers test candidate directions, interpret the evidence, and carry the result…
Knowledge Editing on Black-box Large Language Models
Xiaoshuai Song, Zhengyang Wang, Keqing He +4
Knowledge editing (KE) aims to efficiently and precisely modify the behavior of large language models (LLMs) to update specific knowledge without negatively influencing other knowl…
ToolScope: An Agentic Framework for Vision-Guided and Long-Horizon Tool Use
Mengjie Deng, Guanting Dong, Zhicheng Dou
Recently, large language models (LLMs) have demonstrated remarkable problem-solving capabilities by autonomously integrating with external tools for collaborative reasoning. Howeve…
We-Math: Does Your Large Multimodal Model Achieve Human-like Mathematical Reasoning?
Runqi Qiao, Qiuna Tan, Guanting Dong +15
Visual mathematical reasoning, as a fundamental visual reasoning ability, has received widespread attention from the Large Multimodal Models (LMMs) community. Existing benchmarks,…
InstructERC: Reforming Emotion Recognition in Conversation with Multi-task Retrieval-Augmented Large Language Models
Shanglin Lei, Guanting Dong, Xiaoping Wang +3
The field of emotion recognition of conversation (ERC) has been focusing on separating sentence feature encoding and context modeling, lacking exploration in generative paradigms b…
Generative Zero-Shot Prompt Learning for Cross-Domain Slot Filling with Inverse Prompting
Xuefeng Li, Liwen Wang, Guanting Dong +5
Zero-shot cross-domain slot filling aims to transfer knowledge from the labeled source domain to the unlabeled target domain. Existing models either encode slot descriptions and ex…
CS-Bench: A Comprehensive Benchmark for Large Language Models towards Computer Science Mastery
Xiaoshuai Song, Muxi Diao, Guanting Dong +13
Large language models (LLMs) have demonstrated significant potential in advancing various fields of research and society. However, the current community of LLMs overly focuses on b…
A Prototypical Semantic Decoupling Method via Joint Contrastive Learning for Few-Shot Name Entity Recognition
Guanting Dong, Zechen Wang, Liwen Wang +10
Few-shot named entity recognition (NER) aims at identifying named entities based on only few labeled instances. Most existing prototype-based sequence labeling models tend to memor…
Leveraging LLM-Assisted Query Understanding for Live Retrieval-Augmented Generation
Guanting Dong, Xiaoxi Li, Yuyao Zhang +1
Real-world live retrieval-augmented generation (RAG) systems face significant challenges when processing user queries that are often noisy, ambiguous, and contain multiple intents.…
Toward Effective Tool-Integrated Reasoning via Self-Evolved Preference Learning
Yifei Chen, Guanting Dong, Zhicheng Dou
Tool-Integrated Reasoning (TIR) enables large language models (LLMs) to improve their internal reasoning ability by integrating external tools. However, models employing TIR often…
DotaMath: Decomposition of Thought with Code Assistance and Self-correction for Mathematical Reasoning
Chengpeng Li, Guanting Dong, Mingfeng Xue +3
Large language models (LLMs) have made impressive progress in handling simple math problems, yet they still struggle with more challenging and complex mathematical tasks. In this p…
The Latent Space: Foundation, Evolution, Mechanism, Ability, and Outlook
Xinlei Yu, Zhangquan Chen, Yongbo He +36
Latent space is rapidly emerging as a native substrate for language-based models. While modern systems are still commonly understood through explicit token-level generation, an inc…
Search-o1: Agentic Search-Enhanced Large Reasoning Models
Xiaoxi Li, Guanting Dong, Jiajie Jin +5
Large reasoning models (LRMs) like OpenAI-o1 have demonstrated impressive long stepwise reasoning capabilities through large-scale reinforcement learning. However, their extended r…
CORAL: Benchmarking Multi-turn Conversational Retrieval-Augmentation Generation
Yiruo Cheng, Kelong Mao, Ziliang Zhao +6
Retrieval-Augmented Generation (RAG) has become a powerful paradigm for enhancing large language models (LLMs) through external knowledge retrieval. Despite its widespread attentio…
A Robust Contrastive Alignment Method For Multi-Domain Text Classification
Xuefeng Li, Hao Lei, Liwen Wang +5
Multi-domain text classification can automatically classify texts in various scenarios. Due to the diversity of human languages, texts with the same label in different domains may…
DemoSG: Demonstration-enhanced Schema-guided Generation for Low-resource Event Extraction
Gang Zhao, Xiaocheng Gong, Xinjie Yang +3
Most current Event Extraction (EE) methods focus on the high-resource scenario, which requires a large amount of annotated data and can hardly be applied to low-resource domains. T…
Progressive Multimodal Reasoning via Active Retrieval
Guanting Dong, Chenghao Zhang, Mengjie Deng +3
Multi-step multimodal reasoning tasks pose significant challenges for multimodal large language models (MLLMs), and finding effective ways to enhance their performance in such scen…
DemoNSF: A Multi-task Demonstration-based Generative Framework for Noisy Slot Filling Task
Guanting Dong, Tingfeng Hui, Zhuoma GongQue +5
Recently, prompt-based generative frameworks have shown impressive capabilities in sequence labeling tasks. However, in practical dialogue scenarios, relying solely on simplistic t…
Smaller Language Models Are Better Instruction Evolvers
Tingfeng Hui, Lulu Zhao, Guanting Dong +3
Instruction tuning has been widely used to unleash the complete potential of large language models. Notably, complex and diverse instructions are of significant importance as they…
ChatKBQA: A Generate-then-Retrieve Framework for Knowledge Base Question Answering with Fine-tuned Large Language Models
Haoran Luo, Haihong E, Zichen Tang +9
Knowledge Base Question Answering (KBQA) aims to answer natural language questions over large-scale knowledge bases (KBs), which can be summarized into two crucial steps: knowledge…
Toward General Instruction-Following Alignment for Retrieval-Augmented Generation
Guanting Dong, Xiaoshuai Song, Yutao Zhu +3
Following natural instructions is crucial for the effective application of Retrieval-Augmented Generation (RAG) systems. Despite recent advancements in Large Language Models (LLMs)…
Understand What LLM Needs: Dual Preference Alignment for Retrieval-Augmented Generation
Guanting Dong, Yutao Zhu, Chenghao Zhang +3
Retrieval-augmented generation (RAG) has demonstrated effectiveness in mitigating the hallucination problem of large language models (LLMs). However, the difficulty of aligning the…
Semantic Parsing by Large Language Models for Intricate Updating Strategies of Zero-Shot Dialogue State Tracking
Yuxiang Wu, Guanting Dong, Weiran Xu
Zero-shot Dialogue State Tracking (DST) addresses the challenge of acquiring and annotating task-oriented dialogues, which can be time-consuming and costly. However, DST extends be…
Type-aware Decoding via Explicitly Aggregating Event Information for Document-level Event Extraction
Gang Zhao, Yidong Shi, Shudong Lu +5
Document-level event extraction (DEE) faces two main challenges: arguments-scattering and multi-event. Although previous methods attempt to address these challenges, they overlook…
VeriGraph: Towards Verifiable Data-Analytic Agents
Jiajie Jin, Zhao Yang, Wenle Liao +5
LLM-based agents have demonstrated strong capabilities in data-intensive analytical tasks, yet their outputs are rarely verifiable: a reliance on linear text trajectories makes the…
A Multi-Task Semantic Decomposition Framework with Task-specific Pre-training for Few-Shot NER
Guanting Dong, Zechen Wang, Jinxu Zhao +10
The objective of few-shot named entity recognition is to identify named entities with limited labeled instances. Previous works have primarily focused on optimizing the traditional…
PSSAT: A Perturbed Semantic Structure Awareness Transferring Method for Perturbation-Robust Slot Filling
Guanting Dong, Daichi Guo, Liwen Wang +10
Most existing slot filling models tend to memorize inherent patterns of entities and corresponding contexts from training data. However, these models can lead to system failure or…
We-Math 2.0: A Versatile MathBook System for Incentivizing Visual Mathematical Reasoning
Runqi Qiao, Qiuna Tan, Peiqing Yang +11
Multimodal Large Language Models (MLLMs) have demonstrated impressive capabilities across various tasks, but still struggle with complex mathematical reasoning. Existing research p…
Revisit Input Perturbation Problems for LLMs: A Unified Robustness Evaluation Framework for Noisy Slot Filling Task
Guanting Dong, Jinxu Zhao, Tingfeng Hui +8
With the increasing capabilities of large language models (LLMs), these high-performance models have achieved state-of-the-art results on a wide range of natural language processin…
Scaling Relationship on Learning Mathematical Reasoning with Large Language Models
Zheng Yuan, Hongyi Yuan, Chengpeng Li +5
Mathematical reasoning is a challenging task for large language models (LLMs), while the scaling relationship of it with respect to LLM capacity is under-explored. In this paper, w…
OmniGAIA: Towards Native Omni-Modal AI Agents
Xiaoxi Li, Wenxiang Jiao, Jiarui Jin +10
Human intelligence naturally intertwines omni-modal perception -- spanning vision, audio, and language -- with complex reasoning and tool usage to interact with the world. However,…
Towards Mixed-Modal Retrieval for Universal Retrieval-Augmented Generation
Chenghao Zhang, Guanting Dong, Xinyu Yang +1
Retrieval-Augmented Generation (RAG) has emerged as a powerful paradigm for enhancing large language models (LLMs) by retrieving relevant documents from an external corpus. However…
Tool-Star: Empowering LLM-Brained Multi-Tool Reasoner via Reinforcement Learning
Guanting Dong, Yifei Chen, Xiaoxi Li +7
Recently, large language models (LLMs) have shown remarkable reasoning capabilities via large-scale reinforcement learning (RL). However, leveraging the RL algorithm to empower eff…
Klear-Reasoner: Advancing Reasoning Capability via Gradient-Preserving Clipping Policy Optimization
Zhenpeng Su, Leiyu Pan, Xue Bai +8
We present Klear-Reasoner, a model with long reasoning capabilities that demonstrates careful deliberation during problem solving, achieving outstanding performance across multiple…
MSI-Agent: Incorporating Multi-Scale Insight into Embodied Agents for Superior Planning and Decision-Making
Dayuan Fu, Biqing Qi, Yihuai Gao +3
Long-term memory is significant for agents, in which insights play a crucial role. However, the emergence of irrelevant insight and the lack of general insight can greatly undermin…
Bridging the KB-Text Gap: Leveraging Structured Knowledge-aware Pre-training for KBQA
Guanting Dong, Rumei Li, Sirui Wang +3
Knowledge Base Question Answering (KBQA) aims to answer natural language questions with factual information such as entities and relations in KBs. However, traditional Pre-trained…
Revisiting RAG Ensemble: A Theoretical and Mechanistic Analysis of Multi-RAG System Collaboration
Yifei Chen, Guanting Dong, Yutao Zhu +1
Retrieval-Augmented Generation (RAG) technology has been widely applied in recent years. However, despite the emergence of various RAG frameworks, a single RAG framework still cann…
Watch the Speakers: A Hybrid Continuous Attribution Network for Emotion Recognition in Conversation With Emotion Disentanglement
Shanglin Lei, Xiaoping Wang, Guanting Dong +2
Emotion Recognition in Conversation (ERC) has attracted widespread attention in the natural language processing field due to its enormous potential for practical applications. Exis…