Tianshi Zheng

Jiaxin Bai, Yue Guo, Yifei Dong et al.

Text-agent environments are typically modeled as partially observable Markov decision processes (POMDPs), assuming that the simulator's latent state and transition dynamics are hidden from the agent. Yet little work has examined whether executable code can be induced to serve as a world model for prediction and planning under partial observability. We introduce PatchWorld, a gradient-free framework that turns offline trajectories into executable Python world models through counterexample-guided code repair. Instead of predicting the next observation with a black-box model, PatchWorld induces symbolic belief-state programs whose action updates can be inspected, replayed, and locally patched. Across seven AgentGym environments, PatchWorld-Simple achieves the highest code-based planning score among evaluated methods, reaching 76.4\% macro success in live one-step lookahead while invoking no LLM calls inside the world-model prediction module itself. We further find that a human-specified residual-memory bias improves surface observation fidelity but weakens decision utility. This exposes a tradeoff in executable world models, since improving observation fidelity can come at the expense of action-discriminative dynamics, and vice versa. Code is available at https://github.com/HKBU-KnowComp/PatchWorld.

Yisen Gao, Yixi Cai, Tianshi Zheng et al.

Abductive reasoning over knowledge graphs aims to generate logical hypotheses that explain observed entities or facts. Existing controllable hypothesis generation methods allow users to guide this process with explicit conditions, but they remain limited in interactive settings: they struggle to ground evolving natural-language intents across multi-turn dialogues and provide little fine-grained diagnosis when generated hypotheses fail. To address these limitations, we propose HypoAgent, an Agentic framework for interactive abductive Hypothesis Generation over knowledge graphs. HypoAgent integrates three agents: an Intent Recognition Agent that grounds user utterances and dialogue history into executable KG conditions, a Hypothesis Generation Agent that performs controllable hypothesis generation according to the extracted user intention, and a Root Cause Analysis Agent that diagnoses unreliable hypothesis fragments and leverages KG neighborhood probing to identify supported refinements. Experiments on commonsense and biomedical domain-specific knowledge graphs demonstrate that HypoAgent achieves state-of-the-art semantic similarity under single-turn, multi-turn, and unconditional settings. Our code is available at https://github.com/HKUST-KnowComp/HypoAgent.

Wei Fan, Yining Zhou, Mufan Zhang et al.

While large language models (LLMs) augmented with agentic search capabilities show promise for legal reasoning, they overlook a fundamental constraint that applicable law must match the temporal context of each case, as retroactive application of statutes violates core legal principles and leads to erroneous conclusions. Our observations reveal that current legal LLMs suffer from temporal bias anchored to their training cutoff, while search agents rarely incorporate temporal constraints into queries, and that web search alone cannot provide the precise statute and precedent citations that legal reasoning demands. To address these challenges, we propose LegalSearch-R1, an end-to-end reinforcement learning framework that pairs local statute RAG for precise article matching with online web search for broader legal knowledge, trained on temporally-indexed data spanning multiple amendment periods to enforce temporal consistency. Extensive experiments on our benchmark covering 13 legal tasks demonstrate that our 7B-parameter agent outperforms state-of-the-art deep research frameworks and specialized legal LLMs by 12.9% to 29.8%, surpasses baselines by 57.7% to 80.3% on temporal consistency, and exhibits robust out-of-domain generalization. The code and data are available at https://github.com/AlexFanw/LegalSearch-R1.

Yauwai Yim, Chunkit Chan, Tianyu Shi et al.

Large language models (LLMs) have shown success in handling simple games with imperfect information and enabling multi-agent coordination, but their ability to facilitate practical collaboration against other agents in complex, imperfect information environments, especially in a non-English environment, still needs to be explored. This study investigates the applicability of knowledge acquired by open-source and API-based LLMs to sophisticated text-based games requiring agent collaboration under imperfect information, comparing their performance to established baselines using other types of agents. We propose a Theory of Mind (ToM) planning technique that allows LLM agents to adapt their strategy against various adversaries using only game rules, current state, and historical context as input. An external tool was incorporated to mitigate the challenge of dynamic and extensive action spaces in this card game. Our results show that although a performance gap exists between current LLMs and state-of-the-art reinforcement learning (RL) models, LLMs demonstrate ToM capabilities in this game setting. It consistently improves their performance against opposing agents, suggesting their ability to understand the actions of allies and adversaries and establish collaboration with allies. To encourage further research and understanding, we have made our codebase openly accessible.

Tianshi Zheng, Rui Wang, Xiyun Li et al. · tencent-ai

Frontier scientific reasoning is rapidly emerging as a key foundation for advancing AI agents in automated scientific discovery. Deep research agents offer a promising approach to this challenge. These models develop robust problem-solving capabilities through post-training on information-seeking tasks, which are typically curated via knowledge graph construction or iterative web browsing. However, these strategies face inherent limitations in frontier science, where domain-specific knowledge is scattered across sparse and heterogeneous academic sources, and problem solving requires sophisticated computation and reasoning far beyond factual recall. To bridge this gap, we introduce SciResearcher, a fully automated agentic framework for frontier-science data construction. SciResearcher synthesizes diverse conceptual and computational tasks grounded in academic evidence, while eliciting information acquisition, tool-integrated reasoning, and long-horizon capabilities. Leveraging the curated data for supervised fine-tuning and agentic reinforcement learning, we develop SciResearcher-8B, an agent foundation model that achieves 19.46% on the HLE-Bio/Chem-Gold benchmark, establishing a new state of the art at its parameter scale and surpassing several larger proprietary agents. It further achieves 13-15% absolute gains on SuperGPQA-Hard-Biology and TRQA-Literature benchmarks. Overall, SciResearcher introduces a new paradigm for automated data construction for frontier scientific reasoning and offers a scalable path toward future scientific agents.

Jiaxin Bai, Tianshi Zheng, Yangqiu Song

Complex Query Answering (CQA) is an important and fundamental task for knowledge graph (KG) reasoning. Query encoding (QE) is proposed as a fast and robust solution to CQA. In the encoding process, most existing QE methods first parse the logical query into an executable computational direct-acyclic graph (DAG), then use neural networks to parameterize the operators, and finally, recursively execute these neuralized operators. However, the parameterization-and-execution paradigm may be potentially over-complicated, as it can be structurally simplified by a single neural network encoder. Meanwhile, sequence encoders, like LSTM and Transformer, proved to be effective for encoding semantic graphs in related tasks. Motivated by this, we propose sequential query encoding (SQE) as an alternative to encode queries for CQA. Instead of parameterizing and executing the computational graph, SQE first uses a search-based algorithm to linearize the computational graph to a sequence of tokens and then uses a sequence encoder to compute its vector representation. Then this vector representation is used as a query embedding to retrieve answers from the embedding space according to similarity scores. Despite its simplicity, SQE demonstrates state-of-the-art neural query encoding performance on FB15k, FB15k-237, and NELL on an extended benchmark including twenty-nine types of in-distribution queries. Further experiment shows that SQE also demonstrates comparable knowledge inference capability on out-of-distribution queries, whose query types are not observed during the training process.

Tianshi Zheng, Jiaxin Bai, Yicheng Wang et al. · tencent-ai

While large language models (LLMs) have demonstrated impressive capabilities across various natural language processing tasks by acquiring rich factual knowledge from their broad training data, their ability to synthesize and logically reason with this knowledge in complex ways remains underexplored. In this work, we present a systematic evaluation of state-of-the-art LLMs' complex logical reasoning abilities through a novel benchmark of automatically generated complex reasoning questions over general domain and biomedical knowledge graphs. Our extensive experiments, employing diverse in-context learning techniques, reveal that LLMs excel at reasoning over general world knowledge but face significant challenges with specialized domain-specific knowledge. We find that prompting with explicit Chain-of-Thought demonstrations can substantially improve LLM performance on complex logical reasoning tasks with diverse logical operations. Interestingly, our controlled evaluations uncover an asymmetry where LLMs display proficiency at set union operations, but struggle considerably with set intersections - a key building block of logical reasoning. To foster further work, we will publicly release our evaluation benchmark and code.

Xiyu Ren, Zhaowei Wang, Yiming Du et al.

Memory is essential for large vision-language models (LVLMs) to handle long, multimodal interactions, with two method directions providing this capability: long-context LVLMs and memory-augmented agents. However, no existing benchmark conducts a systematic comparison of the two on questions that genuinely require multimodal evidence. To close this gap, we introduce MEMLENS, a comprehensive benchmark for memory in multimodal multi-session conversations, comprising 789 questions across five memory abilities (information extraction, multi-session reasoning, temporal reasoning, knowledge update, and answer refusal) at four standard context lengths (32K-256K tokens) under a cross-modal token-counting scheme. An image-ablation study confirms that solving MEMLENS requires visual evidence: removing evidence images drops two frontier LVLMs below 2% accuracy on the 80.4% of questions whose evidence includes images. Evaluating 27 LVLMs and 7 memory-augmented agents, we find that long-context LVLMs achieve high short-context accuracy through direct visual grounding but degrade as conversations grow, whereas memory agents are length-stable but lose visual fidelity under storage-time compression. Multi-session reasoning caps most systems below 30%, and neither approach alone solves the task. These results motivate hybrid architectures that combine long-context attention with structured multimodal retrieval. Our code is available at https://github.com/xrenaf/MEMLENS.

Qing Zong, Zhaowei Wang, Baixuan Xu et al.

A main goal of Argument Mining (AM) is to analyze an author's stance. Unlike previous AM datasets focusing only on text, the shared task at the 10th Workshop on Argument Mining introduces a dataset including both text and images. Importantly, these images contain both visual elements and optical characters. Our new framework, TILFA (A Unified Framework for Text, Image, and Layout Fusion in Argument Mining), is designed to handle this mixed data. It excels at not only understanding text but also detecting optical characters and recognizing layout details in images. Our model significantly outperforms existing baselines, earning our team, KnowComp, the 1st place in the leaderboard of Argumentative Stance Classification subtask in this shared task.

Zheye Deng, Chunkit Chan, Weiqi Wang et al.

The task of condensing large chunks of textual information into concise and structured tables has gained attention recently due to the emergence of Large Language Models (LLMs) and their potential benefit for downstream tasks, such as text summarization and text mining. Previous approaches often generate tables that directly replicate information from the text, limiting their applicability in broader contexts, as text-to-table generation in real-life scenarios necessitates information extraction, reasoning, and integration. However, there is a lack of both datasets and methodologies towards this task. In this paper, we introduce LiveSum, a new benchmark dataset created for generating summary tables of competitions based on real-time commentary texts. We evaluate the performances of state-of-the-art LLMs on this task in both fine-tuning and zero-shot settings, and additionally propose a novel pipeline called $T^3$(Text-Tuple-Table) to improve their performances. Extensive experimental results demonstrate that LLMs still struggle with this task even after fine-tuning, while our approach can offer substantial performance gains without explicit training. Further analyses demonstrate that our method exhibits strong generalization abilities, surpassing previous approaches on several other text-to-table datasets. Our code and data can be found at https://github.com/HKUST-KnowComp/LiveSum.

Jiayu Liu, Rui Wang, Qing Zong et al.

Accurately assessing model confidence is essential for deploying large language models (LLMs) in mission-critical factual domains. While retrieval-augmented generation (RAG) is widely adopted to improve grounding, confidence calibration in RAG settings remains poorly understood. We conduct a systematic study across four benchmarks, revealing that LLMs exhibit poor calibration performance due to noisy retrieved contexts. Specifically, contradictory or irrelevant evidence tends to inflate the model's false certainty, leading to severe overconfidence. To address this, we propose NAACL Rules (Noise-AwAre Confidence CaLibration Rules) to provide a principled foundation for resolving overconfidence under noise. We further design NAACL, a noise-aware calibration framework that synthesizes supervision from about 2K HotpotQA examples guided by these rules. By performing supervised fine-tuning (SFT) with this data, NAACL equips models with intrinsic noise awareness without relying on stronger teacher models. Empirical results show that NAACL yields substantial gains, improving ECE scores by 10.9% in-domain and 8.0% out-of-domain. By bridging the gap between retrieval noise and verbal calibration, NAACL paves the way for both accurate and epistemically reliable LLMs.

Tianshi Zheng, Zheye Deng, Hong Ting Tsang et al.

Large Language Models (LLMs) are catalyzing a paradigm shift in scientific discovery, evolving from task-specific automation tools into increasingly autonomous agents and fundamentally redefining research processes and human-AI collaboration. This survey systematically charts this burgeoning field, placing a central focus on the changing roles and escalating capabilities of LLMs in science. Through the lens of the scientific method, we introduce a foundational three-level taxonomy-Tool, Analyst, and Scientist-to delineate their escalating autonomy and evolving responsibilities within the research lifecycle. We further identify pivotal challenges and future research trajectories such as robotic automation, self-improvement, and ethical governance. Overall, this survey provides a conceptual architecture and strategic foresight to navigate and shape the future of AI-driven scientific discovery, fostering both rapid innovation and responsible advancement. Github Repository: https://github.com/HKUST-KnowComp/Awesome-LLM-Scientific-Discovery.

Chunyang Li, Weiqi Wang, Tianshi Zheng et al.

Inductive reasoning, a cornerstone of human cognition, enables generalization from limited data but hasn't yet been fully achieved by large language models (LLMs). While modern LLMs excel at reasoning tasks, their ability to maintain stable and consistent rule abstraction under imperfect observations remains underexplored. To fill this gap, in this work, we introduce Robust Rule Induction, a task that evaluates LLMs' capability in inferring rules from data that are fused with noisy examples. To address this task, we further propose Sample-steered Rule Refinement (SRR), a method enhancing reasoning stability via observation diversification and execution-guided feedback. Experiments across arithmetic, cryptography, and list functions reveal: (1) SRR outperforms other methods with minimal performance degradation under noise; (2) Despite slight accuracy variation, LLMs exhibit instability under noise (e.g., 0% accuracy change with only 70% consistent score); (3) Counterfactual task gaps highlight LLMs' reliance on memorized patterns over genuine abstraction. Our findings challenge LLMs' reasoning robustness, revealing susceptibility to hypothesis drift and pattern overfitting, while providing empirical evidence critical for developing human-like inductive systems. Code and data are available at https://github.com/HKUST-KnowComp/Robust-Rule-Induction.

Haochen Shi, Tianshi Zheng, Weiqi Wang et al.

Large Language Model (LLM) routing is a pivotal technique for navigating a diverse landscape of LLMs, aiming to select the best-performing LLMs tailored to the domains of user queries, while managing computational resources. However, current routing approaches often face limitations in scalability when dealing with a large pool of specialized LLMs, or in their adaptability to extending model scope and evolving capability domains. To overcome those challenges, we propose InferenceDynamics, a flexible and scalable multi-dimensional routing framework by modeling the capability and knowledge of models. We operate it on our comprehensive dataset RouteMix, and demonstrate its effectiveness and generalizability in group-level routing using modern benchmarks including MMLU-Pro, GPQA, BigGenBench, and LiveBench, showcasing its ability to identify and leverage top-performing models for given tasks, leading to superior outcomes with efficient resource utilization. The broader adoption of Inference Dynamics can empower users to harness the full specialized potential of the LLM ecosystem, and our code will be made publicly available to encourage further research.

Rui Wang, Qihan Lin, Jiayu Liu et al.

Prospect Theory (PT) models human decision-making under uncertainty, while epistemic markers (e.g., maybe) serve to express uncertainty in language. However, it remains largely unexplored whether Prospect Theory applies to contemporary Large Language Models and whether epistemic markers, which express human uncertainty, affect their decision-making behaviour. To address these research gaps, we design a three-stage experiment based on economic questionnaires. We propose a more general and precise evaluation framework to model LLMs' decision-making behaviour under PT, introducing uncertainty through the empirical probability values associated with commonly used epistemic markers in comparable contexts. We then incorporate epistemic markers into the evaluation framework based on their corresponding probability values to examine their influence on LLM decision-making behaviours. Our findings suggest that modelling LLMs' decision-making with PT is not consistently reliable, particularly when uncertainty is expressed in diverse linguistic forms. Our code is released in https://github.com/HKUST-KnowComp/MarPT.

Yunxiang Mo, Tianshi Zheng, Qing Zong et al.

Multimodal abductive reasoning--the generation and selection of explanatory hypotheses from partial observations--is a cornerstone of intelligence. Current evaluations of this ability in vision-language models (VLMs) are largely confined to static, single-agent tasks. Inspired by Dixit, we introduce DixitWorld, a comprehensive evaluation suite designed to deconstruct this challenge. DIXITWORLD features two core components: DixitArena, a dynamic, multi-agent environment that evaluates both hypothesis generation (a "storyteller" crafting cryptic clues) and hypothesis selection ("listeners" choosing the target image from decoys) under imperfect information; and DixitBench, a static QA benchmark that isolates the listener's task for efficient, controlled evaluation. Results from DixitArena reveal distinct, role-dependent behaviors: smaller open-source models often excel as creative storytellers, producing imaginative yet less discriminative clues, whereas larger proprietary models demonstrate superior overall performance, particularly as listeners. Performance on DixitBench strongly correlates with listener results in DixitArena, validating it as a reliable proxy for hypothesis selection. Our findings reveal a key trade-off between generative creativity and discriminative understanding in multimodal abductive reasoning, a central challenge for developing more balanced and capable vision-language agents.

Tianshi Zheng, Jiayang Cheng, Chunyang Li et al.

Modern large language models (LLMs) employ diverse logical inference mechanisms for reasoning, making the strategic optimization of these approaches critical for advancing their capabilities. This paper systematically investigate the comparative dynamics of inductive (System 1) versus abductive/deductive (System 2) inference in LLMs. We utilize a controlled analogical reasoning environment, varying modality (textual, visual, symbolic), difficulty, and task format (MCQ / free-text). Our analysis reveals System 2 pipelines generally excel, particularly in visual/symbolic modalities and harder tasks, while System 1 is competitive for textual and easier problems. Crucially, task format significantly influences their relative advantage, with System 1 sometimes outperforming System 2 in free-text rule-execution. These core findings generalize to broader in-context learning. Furthermore, we demonstrate that advanced System 2 strategies like hypothesis selection and iterative refinement can substantially scale LLM reasoning. This study offers foundational insights and actionable guidelines for strategically deploying logical inference to enhance LLM reasoning. Resources are available at https://github.com/HKUST-KnowComp/LogiDynamics.

Yixiang Chen, Tianshi Zheng, Shijue Huang et al.

Test-time scaling without interpreter feedback is essential for real-world code generation scenarios where test cases are not readily available. While existing paradigms often rely on either greedy exploitation (i.e., iterative refinement) or stochastic exploration (i.e., relying on sample-based voting or reranking mechanisms), the balance between these two dimensions remains underexplored. To investigate the LLM's intrinsic ability to balance exploitation and exploration, we introduce SELF-REDRAFT, a framework built upon Self-Refine that encourages the model to propose new drafts for solutions that are fundamentally flawed. Our results show that SELF-REDRAFT consistently achieves better performance than Self-Refine when converged under the same maximum number of iterations. Still, we observe that significant room for improvement remains, largely due to two core aspects of current self-redraft capabilities: constrained capacity for generating instructive feedback and fragile discriminative judgment. We also find that balancing strategies vary notably across different LLMs, reflecting distinct, model-specific behaviors. Overall, our study establishes a baseline for intrinsic exploration-exploitation balancing in test-time scaling and identifies feedback and discrimination as key areas with potential for future advances.

Tianshi Zheng, Yixiang Chen, Chengxi Li et al.

Chain-of-Thought (CoT) prompting has been widely recognized for its ability to enhance reasoning capabilities in large language models (LLMs). However, our study reveals a surprising contradiction to this prevailing perspective within the fundamental domain of pattern-based in-context learning (ICL). Through extensive experiments involving 16 state-of-the-art LLMs and nine diverse pattern-based ICL datasets, we demonstrate that CoT and its reasoning variants consistently underperform direct answering across varying model scales and benchmark complexities. To systematically investigate this unexpected phenomenon, we designed extensive experiments to validate several hypothetical explanations. Our analysis uncovers a fundamental hybrid mechanism of explicit-implicit reasoning driving CoT's performance in pattern-based ICL: while explicit reasoning falters due to LLMs' struggles to infer underlying patterns from demonstrations, implicit reasoning-disrupted by the increased contextual distance of CoT rationales-often compensates, delivering correct answers despite flawed rationales. This hybrid mechanism explains CoT's relative underperformance, as noise from weak explicit inference undermines the process, even as implicit mechanisms partially salvage outcomes. Notably, even long-CoT reasoning models, which excel in abstract and symbolic reasoning, fail to fully overcome these limitations despite higher computational costs. Our findings challenge existing assumptions regarding the universal efficacy of CoT, yielding novel insights into its limitations and guiding future research toward more nuanced and effective reasoning methodologies for LLMs.

Jiaxin Bai, Yicheng Wang, Tianshi Zheng et al.

Abductive reasoning is the process of making educated guesses to provide explanations for observations. Although many applications require the use of knowledge for explanations, the utilization of abductive reasoning in conjunction with structured knowledge, such as a knowledge graph, remains largely unexplored. To fill this gap, this paper introduces the task of complex logical hypothesis generation, as an initial step towards abductive logical reasoning with KG. In this task, we aim to generate a complex logical hypothesis so that it can explain a set of observations. We find that the supervised trained generative model can generate logical hypotheses that are structurally closer to the reference hypothesis. However, when generalized to unseen observations, this training objective does not guarantee better hypothesis generation. To address this, we introduce the Reinforcement Learning from Knowledge Graph (RLF-KG) method, which minimizes differences between observations and conclusions drawn from generated hypotheses according to the KG. Experiments show that, with RLF-KG's assistance, the generated hypotheses provide better explanations, and achieve state-of-the-art results on three widely used KGs.

Qing Zong, Zhaowei Wang, Tianshi Zheng et al.

The rapid development of LLMs has sparked extensive research into their factual knowledge. Current works find that LLMs fall short on questions around low-frequency entities. However, such proofs are unreliable since the questions can differ not only in entity frequency but also in difficulty themselves. So we introduce ComparisonQA benchmark, containing 283K abstract questions, each instantiated by a pair of high-frequency and low-frequency entities. It ensures a controllable comparison to study the role of knowledge frequency in the performance of LLMs. Because the difference between such a pair is only the entity with different frequencies. In addition, we use both correctness and uncertainty to develop a two-round method to evaluate LLMs' knowledge robustness. It aims to avoid possible semantic shortcuts which is a serious problem of current QA study. Experiments reveal that LLMs, including GPT-4o, exhibit particularly low robustness regarding low-frequency knowledge. Besides, we find that uncertainty can be used to effectively identify high-quality and shortcut-free questions while maintaining the data size. Based on this, we propose an automatic method to select such questions to form a subset called ComparisonQA-Hard, containing only hard low-frequency questions.

Jiaxin Bai, Wei Fan, Qi Hu et al.

We present AutoSchemaKG, a framework for fully autonomous knowledge graph construction that eliminates the need for predefined schemas. Our system leverages large language models to simultaneously extract knowledge triples and induce comprehensive schemas directly from text, modeling both entities and events while employing conceptualization to organize instances into semantic categories. Processing over 50 million documents, we construct ATLAS (Automated Triple Linking And Schema induction), a family of knowledge graphs with 900+ million nodes and 5.9 billion edges. This approach outperforms state-of-the-art baselines on multi-hop QA tasks and enhances LLM factuality. Notably, our schema induction achieves 92\% semantic alignment with human-crafted schemas with zero manual intervention, demonstrating that billion-scale knowledge graphs with dynamically induced schemas can effectively complement parametric knowledge in large language models.

Jiaxin Bai, Zihao Wang, Yukun Zhou et al. · tsinghua

Graph databases (GDBs) like Neo4j and TigerGraph excel at handling interconnected data but lack advanced inference capabilities. Neural Graph Databases (NGDBs) address this by integrating Graph Neural Networks (GNNs) for predictive analysis and reasoning over incomplete or noisy data. However, NGDBs rely on predefined queries and lack autonomy and adaptability. This paper introduces Agentic Neural Graph Databases (Agentic NGDBs), which extend NGDBs with three core functionalities: autonomous query construction, neural query execution, and continuous learning. We identify ten key challenges in realizing Agentic NGDBs: semantic unit representation, abductive reasoning, scalable query execution, and integration with foundation models like large language models (LLMs). By addressing these challenges, Agentic NGDBs can enable intelligent, self-improving systems for modern data-driven applications, paving the way for adaptable and autonomous data management solutions.

Tianshi Zheng, Jiazheng Wang, Zihao Wang et al. · tsinghua

Transformers, as the fundamental deep learning architecture, have demonstrated great capability in reasoning. This paper studies the generalizable first-order logical reasoning ability of transformers with their parameterized knowledge and how to improve it. Transformers' capability of first-order reasoning is further captured by whether they can conduct first-order logical entailment, which is quantitatively measured by their performance in answering knowledge graph queries. We establish the connections between (1) two types of distribution shifts studied in out-of-distribution generalization and (2) unseen knowledge and query settings discussed in the task of knowledge graph query answering, which makes it possible to characterize the fine-grained generalizability. Results on our comprehensive dataset showed that transformers \textit{outperform} previous methods designed particularly for this task and provided detailed empirical evidence about the impact of the input query syntax, token embedding, and transformer architectures on their reasoning capability. Interestingly, our results revealed the mismatch of positional encoding and other design choices of transformer architectures in previous practices. Motivated by this, we propose TEGA, a logic-aware architecture that significantly improves the performance in generalizable first-order logical entailment.

Yisen Gao, Jiaxin Bai, Tianshi Zheng et al.

Abductive reasoning in knowledge graphs aims to generate plausible logical hypotheses from observed entities, with broad applications in areas such as clinical diagnosis and scientific discovery. However, due to a lack of controllability, a single observation may yield numerous plausible but redundant or irrelevant hypotheses on large-scale knowledge graphs. To address this limitation, we introduce the task of controllable hypothesis generation to improve the practical utility of abductive reasoning. This task faces two key challenges when controlling for generating long and complex logical hypotheses: hypothesis space collapse and hypothesis oversensitivity. To address these challenges, we propose CtrlHGen, a Controllable logcial Hypothesis Generation framework for abductive reasoning over knowledge graphs, trained in a two-stage paradigm including supervised learning and subsequent reinforcement learning. To mitigate hypothesis space collapse, we design a dataset augmentation strategy based on sub-logical decomposition, enabling the model to learn complex logical structures by leveraging semantic patterns in simpler components. To address hypothesis oversensitivity, we incorporate smoothed semantic rewards including Dice and Overlap scores, and introduce a condition-adherence reward to guide the generation toward user-specified control constraints. Extensive experiments on three benchmark datasets demonstrate that our model not only better adheres to control conditions but also achieves superior semantic similarity performance compared to baselines.

Wei Fan, Tianshi Zheng, Yiran Hu et al.

Legal rules encompass not only codified statutes but also implicit adjudicatory principles derived from precedents that contain discretionary norms, social morality, and policy. While computational legal research has advanced in applying established rules to cases, inducing legal rules from judicial decisions remains understudied, constrained by limitations in model inference efficacy and symbolic reasoning capability. The advent of Large Language Models (LLMs) offers unprecedented opportunities for automating the extraction of such latent principles, yet progress is stymied by the absence of formal task definitions, benchmark datasets, and methodologies. To address this gap, we formalize Legal Rule Induction (LRI) as the task of deriving concise, generalizable doctrinal rules from sets of analogous precedents, distilling their shared preconditions, normative behaviors, and legal consequences. We introduce the first LRI benchmark, comprising 5,121 case sets (38,088 Chinese cases in total) for model tuning and 216 expert-annotated gold test sets. Experimental results reveal that: 1) State-of-the-art LLMs struggle with over-generalization and hallucination; 2) Training on our dataset markedly enhances LLMs capabilities in capturing nuanced rule patterns across similar cases.

Baixuan Xu, Chunyang Li, Weiqi Wang et al.

Designing effective collaboration structure for multi-agent LLM systems to enhance collective reasoning is crucial yet remains under-explored. In this paper, we systematically investigate how collaborative reasoning performance is affected by three key design dimensions: (1) Expertise-Domain Alignment, (2) Collaboration Paradigm (structured workflow vs. diversity-driven integration), and (3) System Scale. Our findings reveal that expertise alignment benefits are highly domain-contingent, proving most effective for contextual reasoning tasks. Furthermore, collaboration focused on integrating diverse knowledge consistently outperforms rigid task decomposition. Finally, we empirically explore the impact of scaling the multi-agent system with expertise specialization and study the computational trade off, highlighting the need for more efficient communication protocol design. This work provides concrete guidelines for configuring specialized multi-agent system and identifies critical architectural trade-offs and bottlenecks for scalable multi-agent reasoning. The code will be made available upon acceptance.

Zheye Deng, Chunkit Chan, Tianshi Zheng et al.

The evolution of AI systems toward agentic operation and context-aware retrieval necessitates transforming unstructured text into structured formats like tables, knowledge graphs, and charts. While such conversions enable critical applications from summarization to data mining, current research lacks a comprehensive synthesis of methodologies, datasets, and metrics. This systematic review examines text-to-structure techniques and the encountered challenges, evaluates current datasets and assessment criteria, and outlines potential directions for future research. We also introduce a universal evaluation framework for structured outputs, establishing text-to-structure as foundational infrastructure for next-generation AI systems.

Tianshi Zheng, Weihan Li, Jiaxin Bai et al.

Retrieval-Augmented Generation (RAG) systems show remarkable potential as question answering tools in the K-12 Education domain, where knowledge is typically queried within the restricted scope of authoritative textbooks. However, discrepancies between these textbooks and the parametric knowledge inherent in Large Language Models (LLMs) can undermine the effectiveness of RAG systems. To systematically investigate RAG system robustness against such knowledge discrepancies, we introduce KnowShiftQA. This novel question answering dataset simulates these discrepancies by applying deliberate hypothetical knowledge updates to both answers and source documents, reflecting how textbook knowledge can shift. KnowShiftQA comprises 3,005 questions across five subjects, designed with a comprehensive question typology focusing on context utilization and knowledge integration. Our extensive experiments on retrieval and question answering performance reveal that most RAG systems suffer a substantial performance drop when faced with these knowledge discrepancies. Furthermore, questions requiring the integration of contextual (textbook) knowledge with parametric (LLM) knowledge pose a significant challenge to current LLMs.

Qing Zong, Jiayu Liu, Tianshi Zheng et al.

Accurate confidence calibration in Large Language Models (LLMs) is critical for safe use in high-stakes domains, where clear verbalized confidence enhances user trust. Traditional methods that mimic reference confidence expressions often fail to capture the reasoning needed for accurate confidence assessment. We propose natural language critiques as a solution, ideally suited for confidence calibration, as precise gold confidence labels are hard to obtain and often require multiple generations. This paper studies how natural language critiques can enhance verbalized confidence, addressing: (1) What to critique: uncertainty (question-focused) or confidence (answer-specific)? Analysis shows confidence suits multiple-choice tasks, while uncertainty excels in open-ended scenarios. (2) How to critique: self-critique or critique calibration training? We propose Self-Critique, enabling LLMs to critique and optimize their confidence beyond mere accuracy, and CritiCal, a novel Critique Calibration training method that leverages natural language critiques to improve confidence calibration, moving beyond direct numerical optimization. Experiments show that CritiCal significantly outperforms Self-Critique and other competitive baselines, even surpassing its teacher model, GPT-4o, in complex reasoning tasks. CritiCal also shows robust generalization in out-of-distribution settings, advancing LLM's reliability.

Tianshi Zheng, Kelvin Kiu-Wai Tam, Newt Hue-Nam K. Nguyen et al. · tencent-ai

Large language models are emerging as powerful tools for scientific law discovery, a foundational challenge in AI-driven science. However, existing benchmarks for this task suffer from a fundamental methodological trilemma, forcing a trade-off between scientific relevance, scalability, and resistance to memorization. Furthermore, they oversimplify discovery as static function fitting, failing to capture the authentic scientific process of uncovering embedded laws through the interactive exploration of complex model systems. To address these critical gaps, we introduce NewtonBench, a benchmark comprising 324 scientific law discovery tasks across 12 physics domains. Our design mitigates the evaluation trilemma by using metaphysical shifts - systematic alterations of canonical laws - to generate a vast suite of problems that are scalable, scientifically relevant, and memorization-resistant. Moreover, we elevate the evaluation from static function fitting to interactive model discovery, requiring agents to experimentally probe simulated complex systems to uncover hidden principles. Our extensive experiment reveals a clear but fragile capability for discovery in frontier LLMs: this ability degrades precipitously with increasing system complexity and exhibits extreme sensitivity to observational noise. Notably, we uncover a paradoxical effect of tool assistance: providing a code interpreter can hinder more capable models by inducing a premature shift from exploration to exploitation, causing them to satisfice on suboptimal solutions. These results demonstrate that robust, generalizable discovery in complex, interactive environments remains the core challenge. By providing a scalable, robust, and scientifically authentic testbed, NewtonBench offers a crucial tool for measuring true progress and guiding the development of next-generation AI agents capable of genuine scientific discovery.

Baixuan Xu, Tianshi Zheng, Zhaowei Wang et al. · tencent-ai

Enabling LLMs to effectively operate long-horizon task which requires long-term planning and multiple interactions is essential for open-world autonomy. Conventional methods adopt planning with actions where a executable action list would be provided as reference. However, this action representation choice would be impractical when the environment action space is combinatorial exploded (e.g., open-ended real world). This naturally leads to a question: As environmental action space scales, what is the optimal action representation for long-horizon agents? In this paper, we systematically study the effectiveness of two different action representations. The first one is conventional planning with actions (PwA) which is predominantly adopted for its effectiveness on existing benchmarks. The other one is planning with schemas (PwS) which instantiate an action schema into action lists (e.g., "move [OBJ] to [OBJ]" -> "move apple to desk") to ensure concise action space and reliable scalability. This alternative is motivated by its alignment with human cognition and its compliance with environment-imposed action format restriction. We propose cognitive bandwidth perspective as a conceptual framework to qualitatively understand the differences between these two action representations and empirically observe a representation-choice inflection point between ALFWorld (~35 actions) and SciWorld (~500 actions), which serve as evidence of the need for scalable representations. We further conduct controlled experiments to study how the location of this inflection point interacts with different model capacities: stronger planning proficiency shifts the inflection rightward, whereas better schema instantiation shifts it leftward. Finally, noting the suboptimal performance of PwS agents, we provide an actionable guide for building more capable PwS agents for better scalable autonomy.

Fangzhou Liang, Tianshi Zheng, Chunkit Chan et al.

Effective multi-agent collaboration requires agents to infer the rationale behind others' actions, a capability rooted in Theory-of-Mind (ToM). While recent Large Language Models (LLMs) excel at logical inference, their ability to infer rationale in dynamic, collaborative settings remains under-explored. This study introduces LLM-Hanabi, a novel benchmark that uses the cooperative game Hanabi to evaluate the rationale inference and ToM of LLMs. Our framework features an automated evaluation system that measures both game performance and ToM proficiency. Across a range of models, we find a significant positive correlation between ToM and in-game success. Notably, first-order ToM (interpreting others' intent) correlates more strongly with performance than second-order ToM (predicting others' interpretations). These findings highlight that for effective AI collaboration, the ability to accurately interpret a partner's rationale is more critical than higher-order reasoning. We conclude that prioritizing first-order ToM is a promising direction for enhancing the collaborative capabilities of future models.

Hong Ting Tsang, Jiaxin Bai, Haoyu Huang et al.

Building effective knowledge graphs (KGs) for Retrieval-Augmented Generation (RAG) is pivotal for advancing question answering (QA) systems. However, its effectiveness is hindered by a fundamental disconnect: the knowledge graph (KG) construction process is decoupled from its downstream application, yielding suboptimal graph structures. To bridge this gap, we introduce AutoGraph-R1, the first framework to directly optimize KG construction for task performance using Reinforcement Learning (RL). AutoGraph-R1 trains an LLM constructor by framing graph generation as a policy learning problem, where the reward is derived from the graph's functional utility in a RAG pipeline. We design two novel, task-aware reward functions, one for graphs as knowledge carriers and another as knowledge indices. Across multiple QA benchmarks, AutoGraph-R1 consistently enables graph RAG methods to achieve significant performance gains over using task-agnostic baseline graphs. Our work shows it is possible to close the loop between construction and application, shifting the paradigm from building intrinsically ``good'' graphs to building demonstrably ``useful'' ones.