AcuityBench Exposes Miscalibrated Clinical Triage Uncertainty
A 914-case benchmark tests 12 models across four care levels, showing that conversational triage trades less over-triage for more under-triage.
Oct 11, 20264 min2605.11398
Computation and Language
cs.CLNatural language processing, computational linguistics, speech, and text retrieval.
Sort:
VoiceChat Preserves Turn Taking With Native Tool Calls
A streaming speech stack separates response text, function calls, transcription, and synthesis, reaching 82.5% tool-selection F1 while maintaining interruption handling.
Oct 4, 20265 min2609.21967
DeepSeek Cuts KV Cache Footprint Fourfold
A causal encoder-decoder MoE pairs cross-layer sparse-attention reuse with FP4 caching to reduce global context memory to 890 bytes per token.
Sep 19, 20264 min2609.19969
Qwen-Audio-3.0-ASR Improves Entity Recall With Hierarchical Hotwords
An instruction-controlled MoE ASR model trained on tens of millions of hours reaches 99.43% recall for priority person names under hotword conditioning.
Sep 12, 20264 min2609.07549
SkillNet Cuts Agent Steps Through Reusable Skills
An ontology, evaluation scheme, and 600,000-skill repository let agents retrieve and compose prior procedures, reporting 40% higher rewards with 30% fewer steps.
Aug 23, 20264 min2603.04448
Agent Swarm Cuts Complex Task Latency Up To 4.5×
Kimi K2.5 couples joint text-vision training with dynamically scheduled parallel subagents, improving agentic search scores while reducing time to target quality.
Aug 13, 20264 min2602.02276
Kimi K3 Brings Open Weights Closer to Frontier Agents
A 2.8T-parameter MoE combines Delta Attention, 16-of-896 expert routing, and agentic reinforcement learning to approach leading proprietary systems at lower task cost.
Aug 10, 20264 min2607.24653
Qwen-CUA Reaches 86.2 on Verified Computer Use
A 397B-A17B mixture-of-experts agent learns screenshot-only keyboard and mouse control from verifiable interactive trajectories, improving OSWorld-Verified performance to 86.2.
Aug 8, 20264 min2608.02352
Mini Activations Narrow Frontier Model Gaps
A sparse 229.9B-parameter MoE activates 9.8B parameters per token and pairs agent-native data with RL for coding, cowork, and reasoning tasks.
Jul 31, 20265 min2605.26494
Hybrid Mamba MoE Trades Dense Scale for Throughput
Soofi S activates 3B of 30B parameters per token and matches larger open models while reaching 4.8k TPS/GPU at 40K context.
Jul 29, 20265 min2607.09424