All entries

Qwen3.8-Flash-Next on 2x3090 + DDR4, part 4: 2.2-2.5x faster prefill by kicking the expert cache off the GPU while the prompt runs

model-releasesr-localllama

Part 4 of the same box. Part 1 was 17 -> 25-29 t/s with the expert cache PR, part 2 was 37-41 t/s after switching to UD-Q4_K_XL and stacking MTP on the cache, part 3 was the top-k fallback that was so

Human

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Hardware

RAGMark: A Comprehensive Framework for Benchmarking Retrieval-Augmented Generation Systems

DGX agent

arXiv:2609.05760v1 Announce Type: cross Abstract: We present RAGMark, a modular benchmarking framework for advanced Retrieval-Augmented Generation (RAG) systems targeting small-scale multi-GPU environ

hardwarearxiv-cs-ai

10 Sept 2026

RAP: Research Attention Prediction Reveals Target-Conditioned Evidence Acquisition Biases

arXiv:2609.10092v1 Announce Type: cross Abstract: Large language models (LLMs) increasingly act as research agents, yet their ability to track shifts in research attention is difficult to evaluate bec

RAPID: Reliability-Aware Pair Importance Distillation

arXiv:2609.05481v1 Announce Type: new Abstract: Inter example relational distillation transfers a teacher's representation geometry by matching relations among examples within a mini batch. Computing

Tutorials

RAU: Reference-based Anatomical Understanding with Vision Language Models

arXiv:2509.22404v2 Announce Type: replace Abstract: Anatomical understanding, which is the ability to identify, localize, or segment anatomical structures, is critical in medical image analysis; howev

tutorialsarxiv-cs-cv

Local Ai

React Native ExecuTorch is now up to 92x faster 🏎️

In v0.10 we achieved significant speedups over v0.9. The video shows the maximum speedups we measured for specific groups of models. Among LLMs, the biggest gain came from Qwen3 0.6B, which runs over

local-air-localllama

Reading a Legal Question Word by Word: Embedding Trajectories of 2,144 Vietnamese Legal Headlines

arXiv:2609.08372v2 Announce Type: replace Abstract: A dense retriever encodes a question as one vector, but the question arrives one word at a time. We read 2,144 held-out headlines from Thu Vien Phap

Real-time and adaptive anomaly detection algorithm for cyclostationary models

arXiv:2609.09326v1 Announce Type: cross Abstract: This article introduces PeriodicCALM, an effective real-time anomaly detection framework designed for cyclostationary data streams. While classical cy

RealSimLoop: Online Real-to-Sim Adaptation via Differentiable Reduced-Order Simulation with Vision Feedback

arXiv:2609.09828v1 Announce Type: cross Abstract: Real-world observations of deformable objects are often sparse or surface-level, while downstream tasks require hidden physical quantities such as int

applicationsarxiv-cs-cv

Reasoning-Aware Compression: Identifying and Protecting Vulnerable Reasoning Circuits for Energy-Efficient LLM Deployment

arXiv:2609.05512v1 Announce Type: new Abstract: Large Reasoning Models (LRMs) impose substantial energy costs during deployment, yet current compression methods apply uniform quantization across all c

Recompilation Is Not Enough: Test-Guided Decompiled-C Repair

arXiv:2609.07201v1 Announce Type: cross Abstract: Decompiled C often becomes recompilable only after repair, but recompilation alone does not establish test-observed behavior. A recompiled command-lin

tutorialsarxiv-cs-ai

Recovering Biomechanical Signals from Missing Keypoints Using Temporal Interpolation in Monocular Gait Analysis

arXiv:2609.09670v1 Announce Type: new Abstract: Monocular pose estimation enables low-cost gait analysis but is sensitive to missing keypoints caused by occlusion, detection errors, or efficiency-driv

Recovering Temporal and Geographic Signals from Language Model Embeddings

arXiv:2609.05721v1 Announce Type: new Abstract: Understanding whether language-model embeddings encode structured real-world information is important for both representation analysis and information r

applicationsarxiv-cs-ai

Recovering topological information of light by topological learning

arXiv:2609.06542v1 Announce Type: cross Abstract: The evolution of modern-day communication networks towards optical solutions with enhanced capacity and robustness is driving interest in topological

RedKnot-MLA: Multi-Head Offline-Online Reuse for DeepSeek-V4 Long-Context Serving

arXiv:2609.07008v1 Announce Type: new Abstract: Multi-head latent attention (MLA) exposes many logical query heads through one packed latent KV stream. This representation is memory efficient, but it

Local Ai

Reduce inference cold starts on Amazon SageMaker HyperPod with model caching

Amazon SageMaker HyperPod now supports model caching for inference, which pre-loads model weights and container images onto cluster nodes so pods read from local NVMe storage instead of downloading ov

local-aiaws-ml-blog

Reduce LLM latency with prefix-aware routing on Amazon SageMaker Inference

model-releasesaws-ml-blog

Amazon SageMaker Inference now offers prefix-aware routing, a routing strategy that sends requests sharing the same prompt prefix to the same instance so the KV cache stays warm. In benchmarks on Llam

Reducing Hallucinations in LLM-based Scientific Literature Analysis Using Peer Context Outlier Detection

arXiv:2604.01461v2 Announce Type: replace Abstract: Reducing hallucinations in Large Language Models (LLMs) is essential for accurate data extraction from large text corpora. Current methods, like pro

Regularized Estimation and Feature Selection in Mixtures of Generalized Linear Experts

model-releasesarxiv-cs-lg

arXiv:1907.06994v2 Announce Type: replace-cross Abstract: Mixtures of experts (MoE) are conditional mixture models in which both the mixing proportions and the component densities depend on the predic

Tools

Reinforcement learning at @cognition's scale is a hard infrastructure problem. We are proud to be part of the stack behind it. Congrats to t…

Reinforcement learning at @cognition's scale is a hard infrastructure problem. We are proud to be part of the stack behind it. Congrats to the team on SWE-2! Read more about how we think about RL at F

toolsfireworks-ai--x

Reliability-Aware Hybrid-K Ensemble Selection for Cervical Cytology Classification: Integrating Discrimination, Calibration, and Selective Prediction

arXiv:2609.09189v1 Announce Type: cross Abstract: High classification accuracy alone is insufficient for clinical image analysis, where calibrated confidence and reliable uncertainty estimates are ess

Reproducing Omitted Temporal Expressions in Japanese News for Retrieval-Augmented Applications

arXiv:2609.09569v1 Announce Type: new Abstract: News articles often contain omitted temporal expressions, such as day-only or month-only mentions, which must be interpreted with reference to the publi

ResidualAuth: What Authorization State Must Language Agents Preserve under Revocable Delegation?

arXiv:2609.08062v1 Announce Type: new Abstract: Tool-using language agents can delegate and revoke permissions while acting through external services. We show that two authorization histories can have

agentsarxiv-cs-ai

Rethinking Sign Language Translation: The Impact of Signer Dependence on Model Evaluation

arXiv:2609.07965v1 Announce Type: cross Abstract: Sign Language Translation has advanced with deep learning, yet evaluations remain largely signer-dependent, with overlapping signers across train/dev/

Retrofitting Code Using LLMs to Support Exceptional Behavior

arXiv:2609.10397v1 Announce Type: cross Abstract: Exception Related Code (ERC), which includes throw statements, conditions (if statements) that guard those throw statements, and try/catch blocks, is

Revisiting the Shape Convention of Transformer Language Models

arXiv:2602.06471v2 Announce Type: replace Abstract: The architectural shape of dense Transformers has remained remarkably stable: narrow-wide-narrow feed-forward networks (FFNs) consume most non-embed

Safety

Revoked but Still Authoritative: An Empirical Study of Revocation Enforcement in Agent-Memory Systems

arXiv:2609.08258v1 Announce Type: new Abstract: Long-running language-model agents depend on persistent memory. Many agent-memory systems preserve history through soft revocation: a contradicted fact

safetyarxiv-cs-ai

RiLM: Parameter-Efficient Language Modeling via Geodesic Decoding

arXiv:2609.10305v1 Announce Type: new Abstract: Language models under one million parameters matter for edge deployment, domain adaptation, and reproducible research, yet a two-layer LSTM or Transform

Risk-Constrained Belief-Space Optimization for Safe Control under Latent Uncertainty

model-releasesarxiv-cs-ro

arXiv:2604.03868v2 Announce Type: replace-cross Abstract: Many safety-critical control systems operate under latent uncertainty that sensors cannot resolve at decision time. Such uncertainty, arising

Risk Is Not Review Value: Wrong-Answer Exposure Under Bounded Review Budgets

arXiv:2609.07095v1 Announce Type: new Abstract: LLM assistants often produce more answers than humans can review before users see them. Most evaluations ask whether an answer is wrong, unsupported, or

ROAM: Robust Organization of Atomic Memories for Agents through Semantic Relations

arXiv:2609.09778v1 Announce Type: new Abstract: Long-term language-model agents rely on external memory across interactions. Atomic memories are particularly useful: their fine-grained semantic bounda

Safety

RoboDrop: Curating VLA Post-Training Data via Local Gradient Compatibility

arXiv:2609.10021v1 Announce Type: new Abstract: Vision--language--action (VLA) models acquire broad generalization through large-scale pretraining, yet adapting them to a new task and robot embodiment

safetyarxiv-cs-ro

Robust Beam Prediction for V2X Networks with Multi-Modal Sensing

arXiv:2609.10200v1 Announce Type: new Abstract: Integrated sensing and communication (ISAC) provides a promising foundation for beam prediction in future vehicle-to-everything (V2X) networks. However,

Robust Industrial Cyber Physical Classification Using Neuromorphic Temporal Embeddings and Hybrid SNN XGBoost Under Machine Unlearning Attacks

arXiv:2609.09564v1 Announce Type: new Abstract: The digitalisation of electrical distribution networks has increased the exposure of power-grid infrastructure to cyber attacks. Existing intrusion dete

Robustness of shallow graph embedding methods for community detection

arXiv:2405.00636v4 Announce Type: replace-cross Abstract: This study investigates the robustness of shallow graph embedding methods for community detection in the face of network perturbations, specif

RoMa-Omega: What Feed-Forward 3D Models Know About Image Matching

arXiv:2609.09507v1 Announce Type: new Abstract: Learned image matching has experienced significant progress in recent years, culminating in robust and accurate matchers such as RoMa, whose robustness

Rosetta at AlexandriaX-2026: LoRA-Adapted NileChat for Context-Aware Dialectal Arabic Dialogue Translation

arXiv:2609.10395v1 Announce Type: new Abstract: This paper describes the Rosetta system for Subtask 1 (Context-Aware English-to-Dialectal Arabic Dialogue Translation) of the AlexandriaX shared task, p

RouteBridge: Reliability-Routed Bidirectional Distillation Between Neural Radiance Fields and 3D Gaussian Splatting

arXiv:2609.09606v1 Announce Type: new Abstract: Neural radiance fields (NeRFs) and 3D Gaussian Splatting (3DGS) encode a scene with complementary inductive biases, but existing cross-representation di

Router Prior Bias: Preserving Base Routing Structure in MoE Post-Training

arXiv:2609.08115v1 Announce Type: new Abstract: Mixture-of-Experts (MoE) pretraining relies on an auxiliary load-balancing loss (LBL) to drive per-expert utilization toward uniformity. Post-training i

Running Vision Qwen 3.8 27B on a 16GB Card, the config (45tks).

model-releasesr-localllama

I am just sharing my config for Qwen 3.8 27b that fits on a 5060TI, what is cool about this is that you can even get vision! and a 85K context (I have 1.5gb of headroom for more context or a better qu

S^3-Bench: Evaluating Speech Interaction Models as Scientific Voice Assistants

arXiv:2609.09852v1 Announce Type: new Abstract: The advance of multimodal large language models (MLLMs) has fundamentally reshaped the paradigm of human-computer interaction, especially speech interac

SA-Profile: Automated Sulcus Angle Profiling from Super-Resolution MRI

arXiv:2609.10125v1 Announce Type: new Abstract: Trochlear dysplasia (TD) is an abnormality of the femoral trochlea associated with anterior knee pain and patellar instability. The sulcus angle (SA) is

Safe Harness Self-Evolution: A Theoretical Analysis of Feasibility and Limits

arXiv:2609.08175v1 Announce Type: new Abstract: Harness self-evolution is the process by which an agent modifies its prompts, tools, code, or orchestration in response to task feedback while keeping t

agentsarxiv-cs-ai

Safe Learning Under Irreversible Dynamics via Asking for Help

arXiv:2502.14043v3 Announce Type: replace Abstract: Most learning algorithms with formal regret guarantees essentially rely on trying all possible behaviors, which is problematic when some errors cann

agentsarxiv-cs-lg

Safety boundary maintenance in consumer AI systems responding to pediatric health queries: a cross-platform benchmark evaluation under naturalistic and adversarially pressured conditions

arXiv:2601.09721v2 Announce Type: replace-cross Abstract: Consumer artificial intelligence chatbots are now accessed by hundreds of millions of users seeking health information, yet systematic evaluat

SAGE: A Hierarchical Framework for Evaluating Interpretive Literary Quality in Narratives

arXiv:2609.06611v1 Announce Type: cross Abstract: Assessing the literary quality of narratives requires evaluating interpretive dimensions (cultural representation, emotional depth, and philosophical