7354 papers
The paper introduces LaMer, a Meta-RL framework that enhances language agents' ability to actively explore and adapt to novel environments at test time through cross-episode training and in-context policy reflection, significantly outperforming standard RL baselines across diverse tasks.
Re-Depth Anything is a test-time self-supervised framework that enhances monocular depth estimation by fusing foundation models with large-scale 2D diffusion priors to perform label-free refinement via generative re-lighting and Score Distillation Sampling, achieving state-of-the-art results without direct depth tensor optimization.
This paper demonstrates that reasoning Large Language Models significantly reduce cloud query execution costs and data consumption compared to non-reasoning models in Text-to-SQL tasks, while revealing that execution time is a poor proxy for cost efficiency and highlighting the substantial financial risks posed by non-reasoning models' tendency to generate inefficient queries.
This paper introduces NeuroSPICE, a physics-informed neural network framework that solves circuit differential-algebraic equations via backpropagation to provide a flexible alternative to conventional SPICE for simulating emerging nonlinear devices and addressing inverse problems, despite not surpassing traditional solvers in raw speed or accuracy.
This paper introduces the Conservation-Congruent Encoding (CCE) framework, a unified physical theory that defines intelligence as an irreversible process of extracting work while minimizing dissipation, thereby deriving universal computational bounds and linking thermodynamic measurement, quantum decoherence, and spacetime geometry to establish substrate-neutral constraints for both natural and artificial intelligence.
This paper introduces an operator-legible evaluation framework centered on under-prediction risk to demonstrate that standard accuracy metrics fail to capture safety-critical grid forecasting needs, revealing that while explicit weather integration improves reliability, unconstrained probabilistic models often induce "fake safety" through excessive inflation, a problem solved by new Bias/OPR-constrained objectives.
This paper introduces DrivingGen, the first comprehensive benchmark for generative driving world models that addresses the lack of rigorous evaluation by combining a diverse dataset with a novel suite of metrics to assess visual realism, trajectory plausibility, temporal coherence, and controllability, thereby revealing critical trade-offs in current state-of-the-art models.
This paper introduces Batch-of-Thought (BoT), a training-free method that enhances Large Language Model reasoning by jointly processing related queries to leverage cross-instance signals, thereby improving accuracy, calibration, and computational efficiency through a multi-agent reflection architecture.
NC-Bench introduces a theory-grounded benchmark that evaluates the conversational competence of large language models by assessing their ability to manage the form and structure of natural interactions across basic, retrieval-augmented, and complex multi-turn scenarios, revealing that while models excel at basic answering, they struggle significantly with repair and complex sequence management tasks.
This paper audits the LAION-Aesthetics Predictor to reveal how its algorithmic gaze reinforces Western, male, and imperial biases by disproportionately filtering content and prioritizing specific cultural aesthetics, ultimately urging a shift toward pluralistic evaluation methods in AI development.
This paper introduces "Single-Shot Planning," a secure architecture for Computer Use Agents that generates a complete, trusted execution graph before observing untrusted UI states to effectively mitigate prompt injection and branch steering attacks while maintaining competitive task performance.
This paper introduces BoxMind, a closed-loop AI system that transforms unstructured boxing footage into hierarchical tactical indicators and predictive gradients to generate expert-level strategic recommendations, which were validated during the 2024 Paris Olympics by contributing to the Chinese National Team's historic medal success.
This paper proposes the Multifaceted Scenario-Aware Hypergraph Learning (MSAHG) framework, which addresses the limitations of existing methods in handling mobility variations across distinct contexts by constructing scenario-specific disentangled sub-hypergraphs and employing a parameter-splitting mechanism to resolve inter-scenario conflicts, thereby significantly improving next POI recommendation performance.
DevBench is a realistic, telemetry-driven benchmark comprising 1,800 instances across six languages that evaluates LLMs on code completion tasks with a focus on ecological validity, contamination-free assessment, and detailed diagnostic insights to guide practical model selection and development.
This paper introduces MAS-Orchestra, a training-time framework that optimizes multi-agent system orchestration via function-calling reinforcement learning, alongside the MASBENCH benchmark, to demonstrate that multi-agent benefits are task-dependent and to achieve significant performance gains with over 10x efficiency on complex reasoning tasks.
This paper introduces the Determinism-Faithfulness Assurance Harness (DFAH), a framework and set of financial benchmarks demonstrating that decision determinism and task accuracy in LLM agents are uncorrelated, thereby necessitating independent measurement to ensure reliable regulatory audit replay in financial services.
This paper proposes a novel data-rate-aware continuous-flow architecture for CNN inference on FPGAs that mitigates hardware underutilization caused by data reduction in pooling and strided convolution layers by interleaving signals and sharing resources, thereby enabling the high-throughput implementation of complex models like MobileNet on a single device.
MeanCache is a training-free framework that accelerates Flow Matching inference by replacing instantaneous velocity caching with an average-velocity approach using cached Jacobian-vector products and a trajectory-stability scheduling strategy, achieving significant speedups (up to 4.56X) while maintaining high generation quality across models like FLUX.1 and HunyuanVideo.
This paper introduces BioAgent Bench, a comprehensive evaluation suite and dataset for assessing AI agents in bioinformatics, which reveals that while frontier models can reliably construct multi-step pipelines, they lack robustness against perturbations and may be unsuitable for privacy-sensitive applications compared to open-weight alternatives.
The paper introduces RedSage, an open-source, locally deployable cybersecurity LLM trained on a massive curated dataset and agentic augmentation pipeline, which achieves state-of-the-art performance on specialized cybersecurity benchmarks while also improving general reasoning capabilities.