3836 papers
LLM-FK is a novel multi-agent framework that overcomes the limitations of conventional heuristic and naive LLM methods in detecting foreign keys within large-scale complex databases by coordinating specialized agents to prune the search space, enhance reasoning with domain knowledge, and ensure global schema consistency, thereby achieving superior accuracy and scalability.
This paper presents a novel automated method combining the substitution technique with systematic backtracking and dynamic programming to prove that the bilinear complexity of $3 \times 3\mathbb{F}_2$ is at least 20, thereby improving the previous lower bound of 19.
This empirical study demonstrates that deployment-motivated prompting constraints significantly exacerbate citation hallucinations across four large language models, with no model achieving a citation existence rate above 47.5% and a substantial portion of unverifiable outputs being fabricated, thereby underscoring the critical need for post-hoc verification in academic and software engineering contexts.
This paper introduces BD-VITON, a new benchmark dataset featuring culturally diverse Bangladeshi garments with complex draping and layering challenges, and evaluates the performance of state-of-the-art virtual try-on models on this dataset to demonstrate significant improvements over zero-shot inference.
This paper proposes a TopRank-based coded caching strategy that optimizes delivery rates under non-uniform, unknown file demands by ranking files based on request count differences rather than estimating exact popularities, thereby achieving superior performance and sublinear regret in scenarios with limited users, small cache capacities, or noisy observation data.
This paper introduces MAviS, a domain-adaptive multimodal conversational assistant for avian species that leverages the newly created MAviS-Dataset and is evaluated on the MAviS-Bench to achieve state-of-the-art performance in fine-grained bird species understanding and multimodal question answering.
This paper proposes a modular, cortically inspired architecture for perceptual AI that leverages neuroscientific principles like predictive processing and specialized modules to overcome the interpretability and generalization limitations of current monolithic models, thereby enabling more transparent and human-aligned reasoning.
This paper proposes a training-centered approach that combines adversarial training with a lightweight feature-map smoothing block to generate saliency maps that are simultaneously sparse, input-stable, and output-stable, thereby enhancing their perceived trustworthiness and sufficiency.
This paper introduces Tursio, a secure, on-premises platform that empowers credit unions to query complex structured databases using natural language by leveraging Large Language Models to automatically generate semantic knowledge graphs and compliant query plans.
This study reveals that in factual verification tasks, users' trust and decision-making are primarily driven by the correctness and certainty framing of LLM rationales rather than their presentation format, highlighting the dual potential of well-designed rationales to either support decision-making or miscalibrate trust.
This paper presents a soft-rigid hybrid gripper that utilizes inflatable silicone pockets to actively modulate surface friction via internal air pressure, enabling the secure grasping of diverse objects—from heavy and slippery to fragile items—without relying on excessive normal force.
This empirical study demonstrates that maintaining end-to-end determinism in 5G-TSN networks for IIoT applications requires carefully tuning Time-Aware Shaper (TAS) transmission window offsets based on high-order percentile bounds of 5G downlink delay and jitter to prevent excessive latency or scheduling failures.
This chapter reviews the evolution of data-driven hint generation in intelligent tutoring systems, highlighting how historical student data enables the creation of next-step hints, strategic subgoals, and timely interventions, while also exploring future adaptations involving behavioral data and Large Language Models.
Faster-HEAL is a lightweight, privacy-preserving collaborative perception framework that addresses the challenges of heterogeneous autonomous vehicles by using low-rank visual prompt fine-tuning and pyramid fusion to align diverse features into a unified space, achieving superior detection performance with significantly reduced computational overhead compared to state-of-the-art methods.
FinSheet-Bench introduces a synthetic benchmark modeled on real private equity fund structures to evaluate LLMs on financial spreadsheet tasks, revealing that even the best-performing models currently lack the accuracy required for unsupervised professional use, particularly on complex, large-scale documents, and suggesting that reliable extraction will require separating document understanding from deterministic computation.
This paper presents a computational method for automatically generating mathematical proof exercises with comparable complexity levels by utilizing cut-based tableau proofs and a mechanizable rule extraction procedure to control difficulty for undergraduate discrete mathematics education.
Echo is an advanced agent that leverages code graphs, automatic query refinement, and execution feedback to generate single, high-quality issue reproduction test cases, achieving a new state-of-the-art success rate of 66.28% on the SWT-Bench Verified dataset.
This study demonstrates through numerical simulations of a spatial trust game that inter-role rewards do not linearly promote trust, as excessive rewards can trigger non-return strategies that suppress cooperation, while moderately costly rewards are most effective at consolidating trust clusters.
This paper proposes a self-supervised evolutionary learning framework that extracts individualized neurodynamic progressions and identity manifolds from unlabeled EEG data during safety-critical decision-making, enabling robust user authentication, anomaly detection, and improved generalization without relying on external labels or predefined cognitive models.
This paper introduces VisualScratchpad, an interactive inference-time analysis tool that leverages sparse autoencoders and attention mechanisms to visualize and debug vision language models by linking visual concepts to text tokens, thereby revealing previously underexplored failure modes such as limited cross-modal alignment and misleading visual concepts.