arXiv subjects

Nan Hu

Publications and source records attributed to Nan Hu.

At least 37 records · Page 2Linked to original sources

arXiv2025

MINTQA: A Multi-Hop Question Answering Benchmark for Evaluating LLMs on New and Tail Knowledge

Large language models (LLMs) have demonstrated impressive capabilities in various reasoning tasks but face significant challenges with complex, knowledge-intensive multi-hop queries, particularly those involving new or long-tail knowledge. Existing benchmarks often fail to fully address these challenges. To bridge this gap, we introduce MINTQA (Multi-hop Question Answering on New and Tail Knowledge), a comprehensive benchmark to evaluate LLMs' capabilities in multi-hop reasoning across four critical dimensions: question handling strategy, sub-question generation, retrieval-augmented generation, and iterative or dynamic decomposition and retrieval. MINTQA comprises 10,479 question-answer pairs for evaluating new knowledge and 17,887 pairs for assessing long-tail knowledge, with each question equipped with corresponding sub-questions and answers. Our systematic evaluation of 22 state-of-the-art LLMs on MINTQA reveals significant limitations in their ability to handle complex knowledge base queries, particularly in handling new or unpopular knowledge. Our findings highlight critical challenges and offer insights for advancing multi-hop reasoning capabilities. The MINTQA benchmark is available at https://github.com/probe2/multi-hop/.

cs.CL↗

arXiv2025

RAG Safety: Exploring Knowledge Poisoning Attacks to Retrieval-Augmented Generation

Retrieval-Augmented Generation (RAG) enhances large language models (LLMs) by retrieving external data to mitigate hallucinations and outdated knowledge issues. Benefiting from the strong ability in facilitating diverse data sources and supporting faithful reasoning, knowledge graphs (KGs) have been increasingly adopted in RAG systems, giving rise to KG-based RAG (KG-RAG) methods. Though RAG systems are widely applied in various applications, recent studies have also revealed its vulnerabilities to data poisoning attacks, where malicious information injected into external knowledge sources can mislead the system into producing incorrect or harmful responses. However, these studies focus exclusively on RAG systems using unstructured textual data sources, leaving the security risks of KG-RAG largely unexplored, despite the fact that KGs present unique vulnerabilities due to their structured and editable nature. In this work, we conduct the first systematic investigation of the security issue of KG-RAG methods through data poisoning attacks. To this end, we introduce a practical, stealthy attack setting that aligns with real-world implementation. We propose an attack strategy that first identifies adversarial target answers and then inserts perturbation triples to complete misleading inference chains in the KG, increasing the likelihood that KG-RAG methods retrieve and rely on these perturbations during generation. Through extensive experiments on two benchmarks and four recent KG-RAG methods, our attack strategy demonstrates strong effectiveness in degrading KG-RAG performance, even with minimal KG perturbations. In-depth analyses are also conducted to understand the safety threats within the internal stages of KG-RAG systems and to explore the robustness of LLMs against adversarial knowledge.

cs.CR↗

arXiv2025

Can LLMs Evaluate Complex Attribution in QA? Automatic Benchmarking using Knowledge Graphs

Attributed Question Answering (AQA) has attracted wide attention, but there are still several limitations in evaluating the attributions, including lacking fine-grained attribution categories, relying on manual annotations, and failing to compare attributions with only subtle differences. To bridge these gaps, we introduce Complex Attributed Question Answering (CAQA), a large-scale benchmark containing comprehensive attribution categories, automatically generated using Knowledge Graphs (KGs), and complex attribution scenarios. We have conducted extensive experiments to verify the effectiveness of CAQA, including the benchmarking of 25 automatic evaluators, their comparison with human evaluators, the testing of LLM evaluators fine-tuned by CAQA and so on. These experiments also lead to a series of important findings that can benefit the future research of AQA. All the codes and data are publicly accessible at https://github.com/HuuuNan/CAQA-Benchmark.

cs.CL↗

arXiv2025

MiniMax-M1: Scaling Test-Time Compute Efficiently with Lightning Attention

We introduce MiniMax-M1, the world's first open-weight, large-scale hybrid-attention reasoning model. MiniMax-M1 is powered by a hybrid Mixture-of-Experts (MoE) architecture combined with a lightning attention mechanism. The model is developed based on our previous MiniMax-Text-01 model, which contains a total of 456 billion parameters with 45.9 billion parameters activated per token. The M1 model natively supports a context length of 1 million tokens, 8x the context size of DeepSeek R1. Furthermore, the lightning attention mechanism in MiniMax-M1 enables efficient scaling of test-time compute. These properties make M1 particularly suitable for complex tasks that require processing long inputs and thinking extensively. MiniMax-M1 is trained using large-scale reinforcement learning (RL) on diverse problems including sandbox-based, real-world software engineering environments. In addition to M1's inherent efficiency advantage for RL training, we propose CISPO, a novel RL algorithm to further enhance RL efficiency. CISPO clips importance sampling weights rather than token updates, outperforming other competitive RL variants. Combining hybrid-attention and CISPO enables MiniMax-M1's full RL training on 512 H800 GPUs to complete in only three weeks, with a rental cost of just $534,700. We release two versions of MiniMax-M1 models with 40K and 80K thinking budgets respectively, where the 40K model represents an intermediate phase of the 80K training. Experiments on standard benchmarks show that our models are comparable or superior to strong open-weight models such as the original DeepSeek-R1 and Qwen3-235B, with particular strengths in complex software engineering, tool utilization, and long-context tasks. We publicly release MiniMax-M1 at https://github.com/MiniMax-AI/MiniMax-M1.

cs.CL↗

arXiv2025

OneEval: Benchmarking LLM Knowledge-intensive Reasoning over Diverse Knowledge Bases

Large Language Models (LLMs) have demonstrated substantial progress on reasoning tasks involving unstructured text, yet their capabilities significantly deteriorate when reasoning requires integrating structured external knowledge such as knowledge graphs, code snippets, or formal logic. This limitation is partly due to the absence of benchmarks capable of systematically evaluating LLM performance across diverse structured knowledge modalities. To address this gap, we introduce \textbf{\textsc{OneEval}}, a comprehensive benchmark explicitly designed to assess the knowledge-intensive reasoning capabilities of LLMs across four structured knowledge modalities, unstructured text, knowledge graphs, code, and formal logic, and five critical domains (general knowledge, government, science, law, and programming). \textsc{OneEval} comprises 4,019 carefully curated instances and includes a challenging subset, \textsc{OneEval}\textsubscript{Hard}, consisting of 1,285 particularly difficult cases. Through extensive evaluation of 18 state-of-the-art open-source and proprietary LLMs, we establish three core findings: a) \emph{persistent limitations in structured reasoning}, with even the strongest model achieving only 32.2\% accuracy on \textsc{OneEval}\textsubscript{Hard}; b) \emph{performance consistently declines as the structural complexity of the knowledge base increases}, with accuracy dropping sharply from 53\% (textual reasoning) to 25\% (formal logic); and c) \emph{diminishing returns from extended reasoning chains}, highlighting the critical need for models to adapt reasoning depth appropriately to task complexity. We release the \textsc{OneEval} datasets, evaluation scripts, and baseline results publicly, accompanied by a leaderboard to facilitate ongoing advancements in structured knowledge reasoning.

cs.CL↗

arXiv2025

Rethinking Stateful Tool Use in Multi-Turn Dialogues: Benchmarks and Challenges

Existing benchmarks that assess Language Models (LMs) as Language Agents (LAs) for tool use primarily focus on stateless, single-turn interactions or partial evaluations, such as tool selection in a single turn, overlooking the inherent stateful nature of interactions in multi-turn applications. To fulfill this gap, we propose \texttt{DialogTool}, a multi-turn dialogue dataset with stateful tool interactions considering the whole life cycle of tool use, across six key tasks in three stages: 1) \textit{tool creation}; 2) \textit{tool utilization}: tool awareness, tool selection, tool execution; and 3) \textit{role-consistent response}: response generation and role play. Furthermore, we build \texttt{VirtualMobile} -- an embodied virtual mobile evaluation environment to simulate API calls and assess the robustness of the created APIs\footnote{We will use tools and APIs alternatively, there are no significant differences between them in this paper.}. Taking advantage of these artifacts, we conduct comprehensive evaluation on 13 distinct open- and closed-source LLMs and provide detailed analysis at each stage, revealing that the existing state-of-the-art LLMs still cannot perform well to use tools over long horizons.

cs.CL↗

arXiv2025

SCoRE: Benchmarking Long-Chain Reasoning in Commonsense Scenarios

Currently, long-chain reasoning remains a key challenge for large language models (LLMs) because natural texts lack sufficient explicit reasoning data. However, existing benchmarks suffer from limitations such as narrow coverage, short reasoning paths, or high construction costs. We introduce SCoRE (Scenario-based Commonsense Reasoning Evaluation), a benchmark that synthesizes multi-hop questions from scenario schemas of entities, relations, and logical rules to assess long-chain commonsense reasoning. SCoRE contains 100k bilingual (Chinese-English) multiple-choice questions whose reasoning chains span 2-11 hops and are grouped into various difficulty levels. Each question is accompanied by fine-grained knowledge labels, explicit reasoning chains, and difficulty levels for diagnostic evaluation. Evaluation results on cutting-edge LLMs such as o3-mini and Deepseek R1 shows that even the best model attains only 69.78% accuracy on SCoRE (even only 47.91% on the hard set), with errors often stemming from rare knowledge, logical inconsistency, and over-interpretation of simple questions. SCoRE offers a scalable, extensible framework for evaluating and diagnosing the long-chain commonsense reasoning abilities of LLMs and guiding future advances in model design and training.

cs.CL↗

arXiv2025

Seed1.5-Thinking: Advancing Superb Reasoning Models with Reinforcement Learning

We introduce Seed1.5-Thinking, capable of reasoning through thinking before responding, resulting in improved performance on a wide range of benchmarks. Seed1.5-Thinking achieves 86.7 on AIME 2024, 55.0 on Codeforces and 77.3 on GPQA, demonstrating excellent reasoning abilities in STEM and coding. Beyond reasoning tasks, the method demonstrates notable generalization across diverse domains. For instance, it surpasses DeepSeek R1 by 8% in win rate on non-reasoning tasks, indicating its broader applicability. Compared to other state-of-the-art reasoning models, Seed1.5-Thinking is a Mixture-of-Experts (MoE) model with a relatively small size, featuring 20B activated and 200B total parameters. As part of our effort to assess generalized reasoning, we develop two internal benchmarks, BeyondAIME and Codeforces, both of which will be publicly released to support future research. Model trial link: https://www.volcengine.com/experience/ark.

cs.CL↗

arXiv2025

Revealing Actual Viscoelastic Relaxation Times in Capillary Breakup

We use experiments and theory to elucidate the size effect in capillary breakup rheometry, where pre-stretching in the visco-capillary stage causes the apparent relaxation time to be consistently smaller than the actual value. We propose a method accounting for both the experimental size and the finite extensibility of polymers to extract the actual relaxation time. A phase diagram characterizes the expected measurement variability and delineates scaling law conditions. The results refine capillary breakup rheometry for viscoelastic fluids and advance the understanding of breakup dynamics across scales.

cond-mat.soft↗

arXiv2025

Reconfigurable Intelligent Surface (RIS) System Level Simulations for Industry Standards

Reconfigurable intelligent surface (RIS) is an emerging technology for wireless communications. In this paper, extensive system level simulations are conducted for analyzing the performance of multi-RIS and multi-base-station (BS) scenarios, by considering typical settings for industry standards. Pathloss and large-scale fading are taken into account when modeling the RIS cascaded and direct links. The performance metrics considered are the downlink reference signal received power (RSRP) and the signal to interference noise ratio (SINR). The evaluation methodology is compatible with that utilized for technology studies in industry standards development organizations, by considering the uniqueness of RIS. The simulations are comprehensive, and they take into account different layouts of RIS panels and mobiles in a cell, and different densities and sizes of RIS panels. Several practical aspects are considered, including the interference between RIS panels, the phase quantization of RIS elements, and the failure of RIS elements. The impact of near field effects for the RIS-mobile links is analyzed as well. Simulation results demonstrate the potential of RIS-aided deployments in improving the system capacity and cell coverage in 6G mobile systems.

cs.IT↗

arXiv2025

Close-contact melting on hydrophobic textured surfaces: Confinement and meniscus effects

We investigate the dynamics of close-contact melting (CCM) on gas-trapped hydrophobic surfaces, with specific focus on the effects of geometrical confinement and the liquid-air meniscus below the liquid film. By employing dual-series and perturbation methods, we obtain numerical solutions for the effective slip lengths associated with velocity $λ$ and temperature $λ_t$ fields, across various values of aspect ratio $Λ$ (defined as the ratio of the film thickness $h$ to the structure's periodic length $l$) and gas-liquid fraction $ϕ$. Asymptotic solutions of $λ$ and $λ_t$ for $Λ\ll 1$ and $Λ\gg 1$ are derived and summarized for different surface structures, interface shapes and $Λ$, which reveal a different trend for $λ$ and $Λ\ll 1$ and the presence of a meniscus. In the context of constant-pressure CCM, our results indicate that transverse-grooves surfaces consistently reduced the heat transfer. However, longitudinal grooves can enhance heat transfer under the effects of confinement and meniscus when $Λ\lessapprox 0.1$ and $ϕ< 1 - 0.5^{2/3} \approx 0.37$. For gravity-driven CCM, the parameters of $l$ and $ϕ$ determine whether the melting rate is enhanced, reduced, or nearly unaffected. We construct a phase diagram based on the parameter matrix $(\log_{10} l, ϕ)$to delineate these three regimes. Lastly, we derived two asymptotic solutions for predicting the variation in time of the unmelted solid height.

cond-mat.soft↗

arXiv2024

HeGTa: Leveraging Heterogeneous Graph-enhanced Large Language Models for Few-shot Complex Table Understanding

Table understanding (TU) has achieved promising advancements, but it faces the challenges of the scarcity of manually labeled tables and the presence of complex table structures.To address these challenges, we propose HGT, a framework with a heterogeneous graph (HG)-enhanced large language model (LLM) to tackle few-shot TU tasks.It leverages the LLM by aligning the table semantics with the LLM's parametric knowledge through soft prompts and instruction turning and deals with complex tables by a multi-task pre-training scheme involving three novel multi-granularity self-supervised HG pre-training objectives.We empirically demonstrate the effectiveness of HGT, showing that it outperforms the SOTA for few-shot complex TU on several benchmarks.

cs.CL↗

arXiv2024

Trainable Joint Channel Estimation, Detection and Decoding for MIMO URLLC Systems

The receiver design for multi-input multi-output (MIMO) ultra-reliable and low-latency communication (URLLC) systems can be a tough task due to the use of short channel codes and few pilot symbols. Consequently, error propagation can occur in traditional turbo receivers, leading to performance degradation. Moreover, the processing delay induced by information exchange between different modules may also be undesirable for URLLC. To address the issues, we advocate to perform joint channel estimation, detection, and decoding (JCDD) for MIMO URLLC systems encoded by short low-density parity-check (LDPC) codes. Specifically, we develop two novel JCDD problem formulations based on the maximum a posteriori (MAP) criterion for Gaussian MIMO channels and sparse mmWave MIMO channels, respectively, which integrate the pilots, the bit-to-symbol mapping, the LDPC code constraints, as well as the channel statistical information. Both the challenging large-scale non-convex problems are then solved based on the alternating direction method of multipliers (ADMM) algorithms, where closed-form solutions are achieved in each ADMM iteration. Furthermore, two JCDD neural networks, called JCDDNet-G and JCDDNet-S, are built by unfolding the derived ADMM algorithms and introducing trainable parameters. It is interesting to find via simulations that the proposed trainable JCDD receivers can outperform the turbo receivers with affordable computational complexities.

eess.SP↗

arXiv2024

Exploring the Impact of Table-to-Text Methods on Augmenting LLM-based Question Answering with Domain Hybrid Data

Augmenting Large Language Models (LLMs) for Question Answering (QA) with domain specific data has attracted wide attention. However, domain data often exists in a hybrid format, including text and semi-structured tables, posing challenges for the seamless integration of information. Table-to-Text Generation is a promising solution by facilitating the transformation of hybrid data into a uniformly text-formatted corpus. Although this technique has been widely studied by the NLP community, there is currently no comparative analysis on how corpora generated by different table-to-text methods affect the performance of QA systems. In this paper, we address this research gap in two steps. First, we innovatively integrate table-to-text generation into the framework of enhancing LLM-based QA systems with domain hybrid data. Then, we utilize this framework in real-world industrial data to conduct extensive experiments on two types of QA systems (DSFT and RAG frameworks) with four representative methods: Markdown format, Template serialization, TPLM-based method, and LLM-based method. Based on the experimental results, we draw some empirical findings and explore the underlying reasons behind the success of some methods. We hope the findings of this work will provide a valuable reference for the academic and industrial communities in developing robust QA systems.

cs.CL↗

arXiv2024

MLDT: Multi-Level Decomposition for Complex Long-Horizon Robotic Task Planning with Open-Source Large Language Model

In the realm of data-driven AI technology, the application of open-source large language models (LLMs) in robotic task planning represents a significant milestone. Recent robotic task planning methods based on open-source LLMs typically leverage vast task planning datasets to enhance models' planning abilities. While these methods show promise, they struggle with complex long-horizon tasks, which require comprehending more context and generating longer action sequences. This paper addresses this limitation by proposing MLDT, theMulti-Level Decomposition Task planning method. This method innovatively decomposes tasks at the goal-level, task-level, and action-level to mitigate the challenge of complex long-horizon tasks. In order to enhance open-source LLMs' planning abilities, we introduce a goal-sensitive corpus generation method to create high-quality training data and conduct instruction tuning on the generated corpus. Since the complexity of the existing datasets is not high enough, we construct a more challenging dataset, LongTasks, to specifically evaluate planning ability on complex long-horizon tasks. We evaluate our method using various LLMs on four datasets in VirtualHome. Our results demonstrate a significant performance enhancement in robotic task planning, showcasing MLDT's effectiveness in overcoming the limitations of existing methods based on open-source LLMs as well as its practicality in complex, real-world scenarios.

cs.RO↗

arXiv2024

MIKE: A New Benchmark for Fine-grained Multimodal Entity Knowledge Editing

Multimodal knowledge editing represents a critical advancement in enhancing the capabilities of Multimodal Large Language Models (MLLMs). Despite its potential, current benchmarks predominantly focus on coarse-grained knowledge, leaving the intricacies of fine-grained (FG) multimodal entity knowledge largely unexplored. This gap presents a notable challenge, as FG entity recognition is pivotal for the practical deployment and effectiveness of MLLMs in diverse real-world scenarios. To bridge this gap, we introduce MIKE, a comprehensive benchmark and dataset specifically designed for the FG multimodal entity knowledge editing. MIKE encompasses a suite of tasks tailored to assess different perspectives, including Vanilla Name Answering, Entity-Level Caption, and Complex-Scenario Recognition. In addition, a new form of knowledge editing, Multi-step Editing, is introduced to evaluate the editing efficiency. Through our extensive evaluations, we demonstrate that the current state-of-the-art methods face significant challenges in tackling our proposed benchmark, underscoring the complexity of FG knowledge editing in MLLMs. Our findings spotlight the urgent need for novel approaches in this domain, setting a clear agenda for future research and development efforts within the community.

cs.CL↗

arXiv2023

Retrieve-Rewrite-Answer: A KG-to-Text Enhanced LLMs Framework for Knowledge Graph Question Answering

Despite their competitive performance on knowledge-intensive tasks, large language models (LLMs) still have limitations in memorizing all world knowledge especially long tail knowledge. In this paper, we study the KG-augmented language model approach for solving the knowledge graph question answering (KGQA) task that requires rich world knowledge. Existing work has shown that retrieving KG knowledge to enhance LLMs prompting can significantly improve LLMs performance in KGQA. However, their approaches lack a well-formed verbalization of KG knowledge, i.e., they ignore the gap between KG representations and textual representations. To this end, we propose an answer-sensitive KG-to-Text approach that can transform KG knowledge into well-textualized statements most informative for KGQA. Based on this approach, we propose a KG-to-Text enhanced LLMs framework for solving the KGQA task. Experiments on several KGQA benchmarks show that the proposed KG-to-Text augmented LLMs approach outperforms previous KG-augmented LLMs approaches regarding answer accuracy and usefulness of knowledge statements.

cs.CL↗

arXiv2023

Can ChatGPT Replace Traditional KBQA Models? An In-depth Analysis of the Question Answering Performance of the GPT LLM Family

ChatGPT is a powerful large language model (LLM) that covers knowledge resources such as Wikipedia and supports natural language question answering using its own knowledge. Therefore, there is growing interest in exploring whether ChatGPT can replace traditional knowledge-based question answering (KBQA) models. Although there have been some works analyzing the question answering performance of ChatGPT, there is still a lack of large-scale, comprehensive testing of various types of complex questions to analyze the limitations of the model. In this paper, we present a framework that follows the black-box testing specifications of CheckList proposed by Ribeiro et. al. We evaluate ChatGPT and its family of LLMs on eight real-world KB-based complex question answering datasets, which include six English datasets and two multilingual datasets. The total number of test cases is approximately 190,000. In addition to the GPT family of LLMs, we also evaluate the well-known FLAN-T5 to identify commonalities between the GPT family and other LLMs. The dataset and code are available at https://github.com/tan92hl/Complex-Question-Answering-Evaluation-of-GPT-family.git

cs.CL↗