https://labnotes.tech 2026-04-15T17:30:37.457Z daily 1 https://labnotes.tech/blog/metas-lecun-introduces-sai-a-measurable-alternative-to-agi 2026-03-08T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/8x-terminal-performance-gains-nvidias-data-recipe-lets-32b-beat-480b 2026-02-25T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/mirage-of-synthesis-dreams-agentic-framework-catches-what-static-benchmarks-miss 2026-02-25T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/when-should-ai-agents-ask-for-help-cmu-cowcorpus-maps-four-human-collaboration-styles 2026-02-23T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/gui-owl-1-5-sets-new-bar-for-open-source-gui-agents 2026-02-23T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/unified-latents-hits-1-4-fid-by-replacing-stable-diffusions-ad-hoc-vae-with-a-diffusion-prior 2026-02-23T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/3-5x-faster-image-generation-ddit-dynamically-resizes-patches-in-diffusion-transformers 2026-02-23T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/reasoning-overthinking-solved-sage-cuts-tokens-44-while-improving-accuracy 2026-02-23T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/voice-search-breaks-in-noise-squtr-benchmark-reveals-the-real-bottleneck 2026-02-18T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/even-gpt-5-fails-at-discovery-odysseyarena-exposes-inductive-bottleneck-llm-agents 2026-02-09T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/prompt-fatigue-solved-vibe-aigc-turns-users-into-commanders-of-multi-agent-creative-workflows 2026-02-05T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/baidu-introduces-ernie-5-trillion-parameter-unified-multimodal-moe-rivals-gpt-5 2026-02-05T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/google-introduces-agentic-vision-gemini-3-flash-now-zooms-annotates-and-investigates-images 2026-02-01T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/first-holistic-ocr-model-ocrverse-unifies-document-parsing-and-code-generation 2026-01-30T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/175-faster-prefill-with-better-accuracy-conceptmoe-adaptive-token-compression 2026-01-30T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/260-percent-better-at-catching-moving-objects-dynamicvla-solves-robot-latency-problem 2026-01-30T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/15-hour-agent-runtimes-solved-idea2story-precomputes-research-knowledge-offline 2026-01-30T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/uplift-vs-cross-attention-upsamplers-linear-scaling-meets-sota-quality 2026-01-29T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/2x-faster-vla-inference-shallow-pi-distillation-edge-robotics 2026-01-29T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/6x-fewer-tokens-better-ocr-deepseek-visual-causal-flow 2026-01-29T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/chroma-real-time-voice-cloning-spoken-dialogue 2026-01-27T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/90-percent-attention-sparsity-salad-speeds-up-video-diffusion 2026-01-26T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/73-on-browsecomp-meituans-560b-open-source-model-leads-agentic-benchmarks 2026-01-26T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/fp8-rollout-instability-solved-jet-rl-unifies-precision-for-stable-rl-training 2026-01-26T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/evocua-evolutionary-training-beats-closed-source-computer-use-agents 2026-01-24T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/97ms-first-packet-latency-qwen3-tts-beats-elevenlabs-voice-cloning-10-languages 2026-01-24T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/tsinghua-diffusion-llms-reason-better-without-flexibility 2026-01-24T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/12x-faster-audio-generation-without-discrete-tokens-kyutais-calm-framework 2026-01-23T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/agent-memory-fragmentation-solved-evermemos-achieves-93-on-locomo 2026-01-23T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/simplemem-gives-llm-agents-30x-cheaper-memory-with-26-better-recall 2026-01-23T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/microsofts-agent-lightning-decouples-rl-training-from-agent-logic 2026-01-23T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/18x-faster-audiovisual-generation-lightricks-open-source-ltx-2-rivals-veo-3 2026-01-21T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/agent-memory-loss-solved-infiagent-file-centric-architecture-enables-unlimited-runtime 2026-01-21T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/4-percent-better-accuracy-with-rude-prompts-how-tone-affects-gpt-4o-performance 2026-01-20T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/metas-shaper-2-7x-better-3d-reconstruction-casual-capture 2026-01-20T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/first-cross-universe-character-mixing-mimix-puts-mr-bean-in-tom-and-jerry 2026-01-18T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/deepresearcheval-benchmark-shows-gemini-leads-quality-manus-wins-factual-accuracy 2026-01-17T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/pixel-to-4d-predicts-dynamic-3d-gaussians-one-pass 2026-01-16T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/multiplex-thinking-merges-multiple-paths-math-reasoning 2026-01-16T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/zero-training-data-full-performance-dr-zero-matches-supervised-search-agents 2026-01-16T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/vlm-hallucinations-exposed-vib-probe-pinpoints-and-suppresses-faulty-attention-heads 2026-01-16T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/16x-faster-on-device-video-generation-qualcomms-rehyat-distills-attention-in-160-gpu-hours 2026-01-16T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/training-free-fix-boosts-vision-language-models-3-points-by-correcting-attention-errors 2026-01-16T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/deepseek-v3-2-gold-medal-imo-ioi-matches-gpt-5-open-weights 2026-01-16T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/why-reasoning-models-cheat-on-efficiency-tnt-fix-cuts-tokens-50 2026-01-15T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/why-chain-of-thought-works-researchers-find-a-single-reasoning-switch-in-llms 2026-01-15T00:00:00.000Z monthly 0.8 https://labnotes.tech/blog/why-hyper-connections-explode-at-scale-deepseeks-manifold-fix 2026-01-15T00:00:00.000Z monthly 0.8 https://labnotes.tech/category/ai_agents 2026-04-15T17:30:37.457Z weekly 0.6 https://labnotes.tech/category/llms_language 2026-04-15T17:30:37.457Z weekly 0.6 https://labnotes.tech/category/vision_multimodal 2026-04-15T17:30:37.457Z weekly 0.6 https://labnotes.tech/category/infrastructure 2026-04-15T17:30:37.457Z weekly 0.6 https://labnotes.tech/category/safety_evaluation 2026-04-15T17:30:37.457Z weekly 0.6 https://labnotes.tech/category/voice_ai 2026-04-15T17:30:37.457Z weekly 0.6