Large language model (LLM) agents are increasingly used in trading systems, where model reasoning, tool use, and continual decisions incur costs that are expected to produce trading value. Existing evaluations typically report performance metrics, but rarely examine agentic viability: whether dynamic LLM-mediated decisions convert their induced costs into measurable incremental profit. To apply this criterion, we introduce TradeLens, a trace-grounded diagnostic toolkit for evaluating agentic trading systems from their trading records, runtime traces, and deployment configurations. It reconstructs trading trajectories, attributes profit and cost to interpretable evidence, and diagnoses whether and why an agent pays for its own intelligence. We conduct extensive analysis across backbone models, capital scales, trading frequencies, and system architectures, together with deployment discussion. Our results show that viability hinges on intelligence-to-profit conversion: models exhibit different failure patterns, such as poor asset selection in DeepSeek-V3.2 and negative timing in GLM-4.7, while capital scale, trading frequency, and architecture matter primarily in these runs, especially through their effects on decision-attributed timing value. These findings reframe the evaluation of LLM-based trading agents from capability-centric performance ranking to trace-grounded diagnosis of intelligence-to-profit conversion. Our code is available at https://github.com/ParadooxAI/TradeLens.

Text-to-SQL deployment is moving from one-shot query generation to interactive repair of executable but semantically incorrect SQL. In this setting, a model must revise an erroneous query under public tool feedback while preserving correct input structures. Existing reinforcement-learning signals are ill matched: outcome rewards reveal only whether the final SQL passes, whereas unconstrained process supervision may reward steps not grounded in SQL semantics. We propose SCOPE-SQL, a reinforcement-learning framework that treats SQL structure as an auditable credit space for interactive repair. It maps candidate queries and public feedback into F0–F6 modules—schema/runtime, table/join, projection, predicate, aggregation/grouping, order/limit, and nesting—and assigns action-level credit for repairing target modules, preserving stable modules, gathering evidence, and submitting when ready. Persistent low-credit regions then drive semantic refresh, generating validator-filtered frontier repair tasks. On BIRD-CRITIC-SQLite, SCOPE-SQL with semantic refresh reaches 48.8% success using a Qwen2.5-Coder-7B backbone, improving over current sota RL baselines by 4.4 and 4.2 points. Ablations attribute 4.4 and 1.8 points to process credit and semantic refresh. Trajectory analysis further shows better repair behavior: Repair+Preserve paths rise from 38.8% to 77.5%, while damage-and-recover paths fall from 43.8% to 1.2%. These results suggest that SQL-structured credit is an interpretable training signal for both policy learning and data expansion in interactive SQL repair.

In document AI, reliable analytics require converting long, noisy (often multi-document) corpora into heterogeneous structured data—e.g., tables for numerical fields, graphs for relations, and trees for hierarchies. Yet, one-pass LLM extraction often yields incomplete or inconsistent structures because it lacks explicit verification and opportunities to revise earlier choices. We present CycleIE, an iterative information extraction (IE) framework that closes the loop between reasoning and acting by coupling ReAct with Monte Carlo Tree Search (MCTS). CycleIE employs a multi-agent workflow to iteratively retrieve, structure, extract, and refine content under verification guidance. By treating extraction as a feedback-driven search process, CycleIE systematically corrects omissions and inconsistencies that defeat one-pass methods, while remaining orthogonal to retrieval-augmented generation (RAG). Experiments on challenging document-based QA benchmarks demonstrate that CycleIE delivers >10% relative improvements over strong baselines, particularly in lengthy or multi-document contexts.

Human experts tackle difficult math problems by identifying and executing a few pivotal steps rather than listing every intermediate thought. In contrast, standard Chain-of-Thought (CoT) distillation trains small models on lengthy reasoning traces, encouraging a uniform overthinking style across easy and hard items alike. The result is rigid, slow solutions that sacrifice adaptivity. This approach stands in sharp contrast to human intuition. Humans naturally adapt their problem-solving strategy, dedicating significant effort to difficult problems while finding quick, simple solutions for easier ones. We argue that the root cause lies in the training data: it contains excess information and reasoning steps organized in ways misaligned with human practice. We address this with Difficulty-Aware Distillation(DAD), a procedure for producing training data that mirrors concise human reasoning. A large teacher model first assesses a problem’s difficulty and then rewrites the solution to retain only the essential steps. Using this process, we constructed LiteCoT, a 100,000-example corpus of short, clear rationales, and used it to train our Liter models. With 100k LiteCoT, we outperform models trained on 800k long CoT and cut both training and inference costs. The advantage is consistent across standard math benchmarks, showing that concise, human-aligned data delivers equal or better accuracy with much less compute. For example, on the challenging AIME24 exam, our approach reaches 74.2% Pass@1 using only about 5K inference tokens, surpassing other methods that consume many more tokens.

Long-horizon LLM agents generate traces that could become reusable experience, but raw trajectories are noisy, local, and hard to govern. Agent Skills offer a structured artifact for combining procedural guidance, executable resources, and applicability boundaries. Yet open skill ecosystems contain redundant, uneven, environment-sensitive artifacts, and indiscriminate updates can pollute future context. We present SkillsVote, a lifecycle-governance framework for Agent Skills across collection, recommendation, attribution, and evolution. SkillsVote profiles a million-scale open-source corpus for environment requirements, quality, and verifiability, and synthesizes tasks for verifiable skills. Before execution, it performs agentic library search over structured skill folders to expose instructional context. After execution, it decomposes trajectories into skill-linked subtasks, attributes outcomes to skill-guided execution, agent exploration, environment, and result signals, and admits only successful reusable discoveries to evidence-gated updates. Experiments on Terminal-Bench 2.0 and SWE-Bench Pro show that SkillsVote improves agent performance on challenging agentic coding benchmarks. The gains arise from two complementary pathways: online evolution over test-time task streams and offline transfer via frozen libraries built from either historical trajectories or curated open-source skills. We release our code at https://github.com/MemTensor/skills-vote.

Text-to-chart retrieval enables users to locate relevant charts through natural language queries and has emerged as a critical capability in modern business intelligence systems. However, current benchmarks fail to simulate realistic user queries or test deep semantic understanding, leaving a significant evaluation gap. To address this, we introduce CRBench, the first real-world BI-sourced benchmark comprising 21,862 charts and 326 queries, which adopts a Target-and-Distractor paradigm to evaluate discriminative retrieval among highly similar candidates. Benchmarking reveals that existing methods, which rely primarily on visual features, fail to capture the rich analytical semantics of charts. We therefore propose a semantic insights synthesis pipeline that automatically generates three hierarchical levels of insights for alignment training: visual patterns, statistical properties, and practical applications. Using this pipeline, we produce 207,498 insights for 69,166 charts as training data. Leveraging this data, we develop ChartFinder, which significantly outperforms state-of-the-art methods on CRBench, achieving 66.9% NDCG@10 for precise queries (an 11.58% improvement) and an average gain of 5% across fuzzy query metrics.

Multi-vector visual document retrieval models encode pages into hundreds of patch-level embeddings, enabling fine-grained interaction but incurring substantial storage overhead. Existing train-free compression approaches rely on token pruning or heuristic merging, which often fail to preserve the broad semantic coverage of the original representations under constrained budgets. To address this, we propose CovMerge, a geometry-aware, train-free index compression framework that formulates token reduction as a budgeted semantic coverage problem on the unit hypersphere. Given frozen patch embeddings, CovMerge applies Farthest Point Sampling (FPS) for coverage-aware prototype initialization, followed by spherical K-Means refinement aligned with MaxSim retrieval. The primary variant, Global Coverage, operates directly on all page patches without external dependencies, providing the strongest simplicity–efficiency trade-off. An optional Regional Coverage extension leverages document layout structure to guide prototype allocation, offering additional gains on layout-sensitive documents when reliable detection is available. Operating purely as post-processing without retraining, CovMerge supports flexible deployment budgets. On the ViDoRe V1 benchmark, CovMerge retains up to 98.1% and 98.6% of the original nDCG@5 for ColPali and ColQwen2, respectively, while using only 10% of the original storage, establishing a new state-of-the-art for storage-efficient multimodal retrieval.

As video large language models (Video-LLMs) become increasingly integrated into real-world applications that demand grounded multimodal reasoning, ensuring their factual consistency and reliability is of critical importance. However, sycophancy, the tendency of these models to align with user input even when it contradicts the visual evidence, undermines their trustworthiness in such contexts. Current sycophancy research has largely overlooked its specific manifestations in the video-language domain, resulting in a notable absence of systematic benchmarks and targeted evaluations to understand how Video-LLMs respond under misleading user input. To fill this gap, we propose ViSE (Video-LLM Sycophancy Benchmarking and Evaluation), the first benchmark designed to evaluate sycophantic behavior in state-of-the-art Video-LLMs across diverse question formats, prompt biases, and visual reasoning tasks. Specifically, ViSE pioneeringly brings linguistic perspectives on sycophancy into the video domain, enabling fine-grained analysis across multiple sycophancy types and interaction patterns. Furthermore, we propose two potential training-free mitigation strategies revealing potential paths for reducing sycophantic bias: (i) enhancing visual grounding through interpretable key-frame selection and (ii) steering model behavior away from sycophancy via targeted, inference-time intervention on its internal neural representations. Our code is available at https://github.com/William030422/Video-Sycophancy.

While Large Language Models (LLMs) demonstrate impressive proficiency in generating SQL queries, they fundamentally lack the capability to self-evaluate correctness without an execution oracle. This limitation creates a stark Generation-Selection Gap, where high potential accuracy (Pass@K) fails to translate into execution accuracy (Pass@1). Although supervised verifiers offer mitigation, they incur prohibitive annotation costs and suffer from domain fragility. Consequently, recent research has pivoted to the training-free setting. However, existing methods—such as Self-Consistency or LLM-as-a-Judge—remain hampered by systematic bias (consensus on hallucinations) and symbolic blindness (inability to simulate execution states). We introduce DPC (Dual-Paradigm Consistency), a multi-agent framework that reformulates SQL selection from a probabilistic guessing task on hidden data into a deterministic verification task on visible data. Specifically, DPC employs a Slicer and a Tester agent to collaboratively construct a Minimal Distinguishing Database (MDD)—an adversarial, fully observable micro-environment engineered to expose logical discrepancies between candidates. To break the self-correction bias, a Solver agent then verifies the SQL candidates by cross-referencing their execution against a parallel Python/Pandas solution. By validating execution consistency between declarative (SQL) and imperative (Python) paradigms, DPC robustly discriminates correct logic from systematic hallucinations. Experiments on BIRD and Spider across multiple LLMs demonstrate that our method consistently outperforms existing selection baselines, achieving absolute accuracy improvements of up to 2.2% over strong competitors like Self-Consistency.

Execution Accuracy (EX), the widely used metric for evaluating the effectiveness of Natural Language to SQL (NL2SQL) solutions, is becoming increasingly unreliable. It is sensitive to syntactic variation, ignores that questions may admit multiple interpretations, and is easily misled by erroneous ground-truth SQL. To address this, we introduce ROSE, an intent-centered metric that focuses on whether the predicted SQL answers the question, rather than consistency with the ground-truth SQL under the reference-dependent paradigm. ROSE employs an adversarial Prover-Refuter cascade: SQL Prover assesses the semantic correctness of a predicted SQL against the user’s intent independently, while Adversarial Refuter uses the ground-truth SQL as evidence to challenge and refine this judgment. On our expert-aligned validation set ROSE-VEC, ROSE achieves the best agreement with human experts, outperforming the next-best metric by nearly 24% in Cohen’s Kappa. We also conduct a large-scale re-evaluation of 19 NL2SQL methods, revealing four valuable insights. We release ROSE and ROSE-VEC to facilitate more reliable NL2SQL research.

Infographics are composite visual artifacts that combine data visualizations with textual and illustrative elements to communicate information. While recent text-to-image (T2I) models can generate aesthetically appealing images, their reliability in generating infographics remains unclear. Generated infographics may appear correct at first glance but contain easily overlooked issues, such as distorted data encoding or incorrect textual content. We present IGenBench, the first benchmark for evaluating the reliability of text-to-infographic generation, comprising 600 curated test cases spanning 30 infographic types. We design an automated evaluation framework that decomposes reliability verification into atomic yes/no questions based on a taxonomy of 10 question types. We employ multimodal large language models (MLLMs) to verify each question, yielding question-level accuracy (Q-ACC) and infographic-level accuracy (I-ACC). We comprehensively evaluate 10 state-of-the-art T2I models on IGenBench. Our systematic analysis reveals key insights for future model development: (i) a three-tier performance hierarchy with the top model achieving Q-ACC of 0.90 but I-ACC of only 0.49; (ii) data-related dimensions emerging as universal bottlenecks (e.g., Data Completeness: 0.21); and (iii) the challenge of achieving end-to-end correctness across all models.

Well-designed prompts are crucial for enhancing Large language models’ (LLMs) reasoning capabilities while aligning their outputs with task requirements across diverse domains. However, manually designed prompts require expertise and iterative experimentation. While existing prompt optimization methods aim to automate this process, they rely heavily on external references such as ground truth or by humans, limiting their applicability in real-world scenarios where such data is unavailable or costly to obtain. To address this, we propose Self-Supervised Prompt Optimization (SPO), a cost-efficient framework that discovers effective prompts for both closed and open-ended tasks without requiring external reference. Motivated by the observations that prompt quality manifests directly in LLM outputs and LLMs can effectively assess adherence to task requirements, we derive evaluation and optimization signals purely from output comparisons. Specifically, SPO selects superior prompts through pairwise output comparisons evaluated by an LLM evaluator, followed by an LLM optimizer that aligns outputs with task requirements. Extensive experiments demonstrate that SPO outperforms state-of-the-art prompt optimization methods, achieving comparable or superior results with significantly lower costs (e.g., 1.1% to 5.6% of existing methods) and fewer samples (e.g., three samples).

Large Language Model (LLM)-based agents have excelled in various domains but face significant challenges when applied to data science workflows due to their complex, multi-stage nature. Current LLM-based agents struggle with non-linear relationships, recursive dependencies, implicit data- and logic-dependent reasoning, and managing extensive context. In this paper, we introduce Data Interpreter, an LLM-based agent that addresses these challenges through hierarchical graph-based modeling to represent the complexity and a progressive strategy for step-by-step verification, refinement, and consistent context management. Extensive experiments confirm the effectiveness of Data Interpreter. On InfiAgent-DABench, it boosts performance by 25% (from 75.9% to 94.9%), and on machine learning and open-ended tasks, it lifts accuracy from 88% to 95% and from 60% to 97%, respectively. Moreover, our method surpasses state-of-the-art baselines by 26% on the MATH dataset. We will release the code upon publication.

Cross-Document Multi-entity question answering (MEQA) demands the integration of scattered information across documents to resolve complex queries involving entities, relationships, and contextual dependencies. Although Large Language Models (LLMs) and Retrieval-augmented Generation (RAG) systems show promise, their performance on cross-document MEQA remains underexplored due to the absence of tailored benchmarks. To address this gap, we introduce MEBench, a scalable multi-document, multi-entity benchmark designed to systematically evaluate LLMs’ capacity to retrieve, consolidate, and reason over scattered and dense information. Our benchmark comprises 4,780 questions which are systematically categorized into three primary categories: Comparative Reasoning, Statistical Reasoning and Relational Reasoning, further divided into eight distinct types, ensuring broad coverage of real-world multi-entity reasoning scenarios. Our experiments on state-of-the-art LLMs reveal critical limitations: even advanced models achieve only 59% accuracy on MEBench. Our benchmark emphasizes the importance of completeness and factual precision of information extraction in MEQA tasks, using Entity-Attributed F1 (EA-F1) metric for granular evaluation of entity-level correctness and attribution validity. MEBench not only highlights systemic weaknesses in current LLM frameworks but also provides a foundation for advancing robust, entity-aware QA architectures.

Chart question answering (ChartQA) tasks play a critical role in interpreting and extracting insights from visualization charts. While recent advancements in multimodal large language models (MLLMs) like GPT-4o have shown promise in high-level ChartQA tasks, such as chart captioning, their effectiveness in low-level ChartQA tasks (e.g., identifying correlations) remains underexplored.In this paper, we address this gap by evaluating MLLMs on low-level ChartQA using a newly curated dataset, ChartInsights, which consists of 22,347 (chart, task, query, answer) covering 10 data analysis tasks across 7 chart types. We systematically evaluate 19 advanced MLLMs, including 12 open-source and 7 closed-source models. The average accuracy rate across these models is 39.8%, with GPT-4o achieving the highest accuracy at 69.17%.To further explore the limitations of MLLMs in low-level ChartQA, we conduct experiments that alter visual elements of charts (e.g., changing color schemes, adding image noise) to assess their impact on the task effectiveness. Furthermore, we propose a new textual prompt strategy, Chain-of-Charts, tailored for low-level ChartQA tasks, which boosts performance by 14.41%, achieving an accuracy of 83.58%. Finally, incorporating a visual prompt strategy that directs attention to relevant visual elements further improves accuracy to 84.32%.

A multimodal large language model MLLMs may struggle with answering visual-based (personal) entity questions (VEQA), such as ”who is A?” or ”who is A that B is talking to?” for various reasons, e.g., the absence of the name of A in the caption or the inability of MLLMs to recognize A, particularly for less common entities. Furthermore, even if the MLLMs can identify A, it may refrain from answering due to privacy concerns. In this paper, we introduce a novel method called Matching-Augmented Reasoning (MAR) to enhance VEQA. Given a collection of visual objects with captions, MAR preprocesses each object individually, identifying faces, names, and their alignments within the object. It encodes this information and stores their vector representations in vector databases. When handling VEQA, MAR retrieves matching faces and names and organizes these entities into a matching graph. MAR then derives the answer to the query by reasoning over this matching graph. Extensive experiments show that MAR significantly improves VEQA compared with the state-of-the-art methods using MLLMs.