二〇二六年十月七日
收録 25 篇
- Rethinking Cross-Tokenizer On-Policy Distillation: From Alignment Coverage to Supervision Reliability
- One Model Family, Two Gold-Level Results: Fine-Tuning Nemotron for IOI and IMO
- Introducing Falcon ASR
- yetone/magpie
- RISED: Rubrics for Agentic Multi-Environment Selection and Self-Distillation
- TypeSafe AI scales Jev to trillions of tokens in three days on Modal
- Quoting Jake Boggan
- Remote control for local agents
- Quoting Victoria Kim
- OpenAI “rogue” agent activities found on Wikimedia projects
- llm-openai-decisions 0.1a0
- Sharing AI progress in mathematics
- Advancing computer use with Ironclad
- EmbeddingGemma 2: an open, lightweight multimodal embedding model
- Introducing Mistral Large 4: Le chonk
- EmbeddingGemma 2
- Mistral Large 4
- Using Parseable with Datasette for OpenTelemetry traces
- MiniCPM-V-4.7-35B-A3B
- Expanding our enterprise inference capacity with IBM Cloud and NVIDIA
- The Cyber Risk Discourse is Broken
- transformers v5.19.0
- Scrimshaw Jukebox
- AICR v1.0: Open, stable, and verifiable GPU cluster configuration
- What AI gets wrong and what failure teaches us