arXiv:2608.11698v3 Announce Type: replace-cross Abstract: On-policy distillation (OPD) trains a student on its own trajectories under dense token-level…
ScaleSense: Cost-Intelligent Scaling Framework via Learned Resource Estimation in Alibaba AnalyticDB
arXiv:2608.07945v2 Announce Type: replace-cross Abstract: Cloud-native serverless data warehouses achieve fine-grained elasticity by decoupling storage…
Evidence-Grounded Trustworthy Multimodal Reasoning and Evaluation Benchmark in Complex Urban Scenes
arXiv:2608.10954v2 Announce Type: replace-cross Abstract: While Multimodal Large Language Models (MLLMs) demonstrate impressive performance in benign…
Cohere Releases Parse 5 (parse-v5.0): A 2.3B Vision Language Model That Turns Enterprise Documents Into Markdown
Cohere has released Parse (parse-v5.0), a 2.3B-parameter vision language model that converts PDFs, slides and images into Markdown with HTML tables,…
M-Net: Integrating Spectral Features and Physical Field Operators into Deep Learning for Medical Image Segmentation
arXiv:2608.12196v2 Announce Type: replace-cross Abstract: Purpose: Deep learning-based medical image segmentation has achieved remarkable success, yet…
When Does Latent Communication Pay? A Causal Audit of Relayed KV Caches in Multi-Agent LLMs
arXiv:2608.04893v2 Announce Type: replace-cross Abstract: Multi-agent LLM systems relay key-value caches instead of text and credit their gains to…
REPREC: Representation Driven Parameter-Efficient Recommendation System
arXiv:2607.24845v3 Announce Type: replace-cross Abstract: Large language models (LLMs) have been applied to sequential recommendation by formulating it as…
TriShieldRAG: 3 Rings, One Blind Spot in Layered Defenses for Retrieval-Augmented Generation
arXiv:2607.23838v2 Announce Type: replace-cross Abstract: Retrieval-Augmented Generation (RAG) grounds LLM answers in query-time retrieved documents, so…
Can LVLMs Uncover the Truth Behind Visual Illusions? An Analysis of Perceptual and Reasoning Capabilities
arXiv:2607.27747v2 Announce Type: replace-cross Abstract: Large Vision Language Models have integrated reasoning capabilities, elevating cognitive…
OpenAI cuts off Cursor after SpaceX acquisition, citing Musk’s history of breaking contracts
OpenAI is cutting off the AI coding tool Cursor after SpaceX acquired the company, citing Elon Musk’s track record of breaking contracts. Cursor…
UniVVT: A Unified End-to-End Framework for High-Fidelity Video Virtual Try-on
arXiv:2608.05745v2 Announce Type: replace-cross Abstract: Video Virtual Try-On (VVT) synthesizes a video of a person wearing a target garment while…
Drift-Adaptive ICU Intervention Prediction: Freezing the Physiological Encoder for Auditable Model Updating
arXiv:2607.19020v3 Announce Type: replace-cross Abstract: Clinical decision support degrades as treatment protocols evolve, but the obstacle to updating a…
RePolicy: Reinforcement Learning for Safety-Policy Invocation in Agent Safeguards
arXiv:2608.24275v2 Announce Type: replace Abstract: Safeguarding language model agents requires assessing complete execution trajectories under…
Autoresearch with Coding Agents: Generalizers and Metric-Maximizers on Quran Recitation Data
arXiv:2607.18064v2 Announce Type: replace-cross Abstract: Coding agents can now be left alone to improve software against a score. In this…
ATLAS: Automated Approximation of Transformers for Efficient Homomorphic Inference in One Hour
arXiv:2607.23478v2 Announce Type: replace-cross Abstract: Fully homomorphic encryption (FHE) lets a server run inference on encrypted data with strong…
Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
arXiv:2607.10358v3 Announce Type: replace-cross Abstract: Foundation models are increasingly used as image feature extractors for mammography, but their…
Let Them Steal: Trapping Large Language Model Extraction Attacks with Knowledge Honeypot
arXiv:2606.15810v2 Announce Type: replace-cross Abstract: Large language models deployed as commercial APIs are vulnerable to model extraction attacks,…
Harnessing the Collective Intelligence of AI Agents in the Wild for New Discoveries
arXiv:2606.10402v2 Announce Type: replace-cross Abstract: Scientific discovery is often a collective process: researchers share partial results, inspect…