Adaptive Multi-Token Prediction: How far should we go? It centres on Llama, and also names Meta. Reported by arXiv. Bharat Hunt files it under AI Models — the section covering a new or updated model, its capabilities, benchmarks or availability.
Written by Bharat Hunt from the headline and the coverage below. The original reporting is the source of truth.
Measuring the Serving Stack Instead of the Model: Hidden Confounds in Local Tool-Use Evaluation. It centres on Benchmarks, and also names Ollama and Phi. Reported by arXiv. Bharat Hunt files it under AI Tools and AI Models — the section covering applications people use — assistants, editors, search, productivity.
Principled Under Pressure: Post-Training Decides Whether LLMs Act on Their Own Moral Judgment. It centres on Meta, and also names Llama and Nothing. Reported by arXiv. Bharat Hunt files it under AI Models — the section covering a new or updated model, its capabilities, benchmarks or availability.
Fine-Tuning LLMs for Translation: General Forgetting Mitigation Does Not Preserve MT-Specific Instruction Following. It centres on Fine-tuning, and also names Benchmarks and Llama. Reported by arXiv. Bharat Hunt files it under AI Models and AI Research — the section covering a new or updated model, its capabilities, benchmarks or availability.
Stale-Document Poisoning: When Outdated Retrieval Overrides Correct Model Answers. It centres on RAG, and also names AI Safety and Benchmarks. Reported by arXiv. Bharat Hunt files it under AI Models and AI Research — the section covering a new or updated model, its capabilities, benchmarks or availability.