Translate
Live intelligence

Latest AI updates

567 updates
AllModelsProductsIndustryResearchTutorialOpinion
Research

Context Compression Weakens Refusal Alignment in LLMs

A new study shows that encoder-based context compression systematically weakens refusal behavior in instruction-tuned language models, even when reconstruction fidelity is high. The effect persists across model families and compression architectures, and is traced to attenuation of the decoder's refusal direction. The authors propose Memory Steering, a lightweight inference-time intervention that restores refusal rates without retraining.

arxiv11h agoScore 78
AI safetymodel compressionalignment
Research

LegalEvolver: Rubric-Guided Self-Evolution for Legal Judgment Prediction

LegalEvolver is a query optimization framework for legal judgment prediction that improves retrieval by generating legally informative queries. It uses heuristic seed queries and rubric-guided reinforcement learning to iteratively self-evolve, surpassing initial performance ceilings. Benchmarks show effectiveness over direct SFT+RL and fixed rubrics.

arxiv11h agoScore 78
Legal Judgment PredictionQuery OptimizationSelf-Evolution
Research Editor selected

Process-based catchment model matches deep-learning accuracy across three continents

A new process-based hydrological model, CaRT+, built on physical laws and refined with AI assistance, achieves accuracy comparable to LSTM neural networks on catchments in Britain, the US, and Australia. With 35 parameters per catchment and no daily neural network, it reaches median Nash-Sutcliffe efficiencies of 0.863 and 0.765 in held-out years on British and US data, and outperforms LSTMs on Australian catchments. The results suggest that catchment-scale laws exist as laws of form, with shared constitutive families and only parameters learned.

Why it matters: This is a concrete, recent research result showing a process-based model matching deep learning accuracy, with implications for hydrology and AI-assisted model refinement.

arxiv11h agoScore 85
hydrological modelingdeep learning comparisonphysical laws
Research Editor selected

MA-Mesh: Role-based Self-Healing for Resilient Multi-Agent Collaboration

MA-Mesh is a framework that adds a role-based self-healing layer to multi-agent systems, generating resilience-aware specifications and executable policy code. It introduces cross-agent reflection and role-based failover to improve robustness under uncertainty. Experiments across software development, tool use, and scientific reasoning show significant gains in performance and robustness.

Why it matters: Provides a concrete, novel framework with explicit resilience mechanisms and experimental validation, valuable for AI system design.

arxiv11h agoScore 82
multi-agent systemsresilience engineeringrole-based coordination
Research

GIST: Gibbs-Targeted Hierarchical Data Selection for Offline RLVR

This paper introduces GIST, a two-level data selection method for offline reinforcement learning with verifiable rewards (RLVR). It derives a factorization of the KL divergence between the current policy and the Gibbs-optimal target, identifying which responses and tokens are most informative. GIST selects correct responses on-policy and incorrect responses near the policy's mode, and supervises only the tokens where the mismatch lies. Experiments show accuracy gains across GSM8K, MATH-500, and MBPP for three models, with 66.6% fewer supervised tokens in a controlled comparison.

arxiv11h agoScore 78
Reinforcement LearningData SelectionToken Efficiency
Research Editor selected

Zero-Shot Coordination among LLM Agents

This paper investigates zero-shot coordination (ZSC) in LLM agents, finding that even frontier models struggle to coordinate in simple, controlled environments due to limited understanding of the coordination problem and weak reasoning about partners' beliefs. The authors propose Coordination-friendly definitions (CFDs) as a principled approach to enable robust coordination)Skip

Why it matters: Provides concrete experimental findings and a novel method (CFDs) for improving LLM coordination, relevant to AI research and deployment.

arxiv11h agoScore 82
LLM coordinationzero-shot coordinationmulti-agent systems
Research

Practical Defenses for Face Recognition Pipelines Against Backdoor Attacks

This technical report presents three practical defense tricks against data poisoning-based backdoor attacks in multi-stage face recognition systems, assuming in-house training but compromised third-party data. The defenses include a training-time method to detect poisoned identities and two inference-time filters for suspicious inputscars. Each is evaluated on realistic workloads, aiming to provide actionable hardening techniques for system providers.

arxiv11h agoScore 72
AI securityface recognitionbackdoor defense
Research Editor selected

Robust Closed-Loop LLM Steering Inspired by Neurostimulation

This paper frames LLM steering as a feedback-control problem, using the network's internal state to condition interventions, inspired by closed-loop neurostimulation. It introduces an H-infinity-based controller that explicitly handles model mismatch and disturbances, improving performance over baselines, especially in out-of-distribution settings. The authors also propose a pre-deployment measure, robust steerability, to quantify tolerance to dynamics mismatch.

Why it matters: Offers a novel cross-disciplinary framework with concrete robustness gains and a new evaluation metric, valuable for AI control and safety.

arxiv11h agoScore 82
LLM controlrobustnessfeedback control
Research Editor selected

D3D: Decoupled Diffusion Drafting for Asynchronous Speculative Decoding

D3D is a standalone diffusion drafter that removes inter-round dependency in speculative decoding, allowing drafting of successor blocks during verification in a packed pass. It combines blockwise denoising with auxiliary autoregressive supervision and prunes layers to reduce cost. Across seven benchmarks, D3D achieves 5.00x and 5.71x speedups with Qwen3-8B and Qwen3-32B targets, outperforming DFlash.

Why it matters: This paper presents a concrete, recent AI development with measurable speedup improvements and open-source code, offering practical value for efficient LLM inference.

arxiv11h agoScore 88
Speculative DecodingDiffusion ModelsEfficient Inference
Research

EviSeek: An Evidence-Centric Agentic Framework for Embodied Active Information Seeking

EviSeek is a training-free framework that makes question-derived evidence demands explicit acquisition objectives for embodied active question answering. It links observation-conditioned actions and selectively retained evidence to the information the agent seeks, avoiding undirected interaction. On 3,436 ESI-Bench questions, EviSeek improves accuracy over a Single-model agent by 7.1 percentage points with Qwen3-VL-32B and 3.7 points with Gemini 3.1 Pro, while also improving visual-support-weighted accuracy and reducing token usage.

arxiv11h agoScore 78
Embodied AIActive PerceptionQuestion Answering
Research Editor selected

SumOne: Task-Agnostic Summarizer for Agent Memory Compression

SumOne is a standalone summarizer that compresses long-horizon agent histories via masked reconstruction, using a frozen reconstructor and GRPO to optimize without task-specific rewards. It reduces inference cost by ~78% on DeepSearchQA and ~75% on SWE-bench Verified with minimal performance loss, and transfers across models and harnesses. The method enables plug-and-play context compression for agents.

Why it matters: Provides a concrete, novel AI development with measurable efficiency gains and cross-task transfer, valuable for agent deployment.

arxiv11h agoScore 82
AI agentscontext compactionself-supervised learning
Research Editor selected

AmbiHTMLBench: A Benchmark for HTML-based Visually Rich Document Parsing

AmbiHTMLBench is a new benchmark for evaluating document parsing systems that output HTML instead of Markdown, aiming to preserve fine-grained spatial structure and visual fidelity. It includes 1,115 diverse pages with annotations and uses both objective and subjective evaluation protocols. Tests on ten multimodal systems show that the best objective score is 0.8092, while the subjective winner differs, indicating complementary aspects of parsing quality. Fine-tuning a small expert model shows task-specific adaptation improves document-to-HTML reconstruction.

Why it matters: Introduces a novel benchmark and evaluation methodology for HTML-based document parsing, providing concrete results and insights for researchers and practitioners.

arxiv11h agoScore 80
Document ParsingBenchmarkingMultimodal Systems
Research

Within-Family EfficientNet Scaling under Fixed-Resolution Transfer Learning for SD-OCT Retinal Disease Classification

This study evaluates eight ImageNet-pretrained EfficientNet variants (B0-B7) for four-class retinal disease classification from SD-OCT images at fixed 224x224 resolution. Results show non-monotonic accuracy scaling, with B1 achieving the highest accuracy (96.95%) and Drusen F1, while B7 shows a secondary accuracy recovery. The OCT accuracy ranking differs from ImageNet hierarchy, indicating task-dependent scaling effects.

arxiv11h agoScore 78
EfficientNet scalingtransfer learningOCT classification
Research

Uncertainty-Guided Framework Enhances Microfluidic Droplet Generator Design

Researchers propose a unified methodology combining adaptive surrogate enhancement with simulation-driven inverse design for microfluidic droplet generators. Neural network ensembles predict flow regimes and droplet diameters, with uncertainty quantification guiding targeted simulations to improve accuracy near regime transition boundaries. The framework, validated on independent datasets, nearly doubles predictions within 0-3% error and reduces high-error cases, while inverse design reliably finds physically realizable configurations.

arxiv11h agoScore 78
microfluidic droplet generationsurrogate-based optimizationuncertainty-guided enrichment
Research Editor selected

ArcRAG: Agentic Compositional Retrieval over Knowledge Graphs

ArcRAG introduces a query-time interface induction method that makes retrieval over knowledge graphs portable across different graph constructions. It builds a local graph interface from entities mentioned in a question)Skip planning multi-hop traversals and executing them as Cypher, without needing construction-specific schema or retuning. Experiments on MuSiQue, HotpotQA, and TriviaQA show that this construction-agnostic retriever remains effective across schema-guided and open-information-extraction graphs, though absolute accuracy still depends on construction.

Why it matters: This paper addresses a critical gap in GraphRAG—cross-construction portability—and offers a practical solution with strong experimental validation, valuable for researchers and practitioners.

arxiv11h agoScore 82
GraphRAGretrieval portabilityknowledge graph construction
Research Editor selected

Model Merging Does Not Require Uniform LoRA Ranks

This paper challenges the standard practice of using uniform LoRA ranks across experts in model merging. Experiments with 36 experts across three domains and multiple scales show that assigning per-domain best ranks can improve multi-domain accuracy by up to 1.8 percentage points with averaging and utility-based methods, while magnitude-based conflict-resolution methods suffer. The authors identify that magnitude-based component selection discards useful spectral components and can drop entire experts, causing large performance drops.

Why it matters: Provides concrete evidence and a practical insight that can improve model merging efficiency and accuracy, relevant to AI practitioners.

arxiv11h agoScore 85
model mergingLoRAmulti-task learning
Research

FPTR: Feasibility-Preserving Transactional Refinement for Deadline-Constrained Joint Beam and Resource Allocation

FPTR is a new scheduling method for wireless systems that always maintains a feasible allocation (incumbent) and refines it in stages, only replacing it if a candidate is complete, validated, and improves the objective. On synthetic benchmarks, the full pipeline improves traffic by 3.83% over a lightweight reference and outperforms six search baselines by 6.68-8.35%, with all runs meeting the 100 ms deadline. The approach separates mandatory output construction from optional refinement, enabling better use of spare computation time.

arxiv11h agoScore 78
wireless schedulingoptimizationreal-time systems
Research Editor selected

Coca-Cola: Latent Context Compression for LLM Agents

Coca-Cola compresses tool observations into latent tokens embedded in the agent's input space, reducing token consumption by 70-89% and runtime by 24-56% while improving task performance on QA benchmarks. Action-Preserving Training aligns next-action distributions to retain decision-critical information.

Why it matters: Offers a concrete, novel method for agent context compression with significant efficiency gains and open-source code.

arxiv11h agoScore 88
LLM agentscontext compressionlatent representation