TD Stuff

All Agentic Systems Creative Code Damage Control Products & Announcements Programming Under the Hood

Topic: Model Fine-Tuning

Autoresearch: Autonomous AI Agents for Self-Improving LLMs

Mar 8, 2026201

An autonomous framework where AI agents independently iterate on and optimize LLM training code within fixed time budgets.

AI Agents Self-Modifying AI LLM Training AI for Science Model Fine-Tuning

Products & Announcements

SERA: Open, Low‑Cost, Repo‑Adaptive Coding Agents

Jan 27, 2026253

SERA makes strong, repo-adaptive coding agents cheap, open, and easy by replacing complex RL with soft-verified, workflow-faithful SFT.

AI Coding Agents Open Source Model Fine-Tuning AI Benchmarks

Under the Hood

Anthropic Confirms Claude 4.5 ‘Soul Doc’ Training, Tied to Better Prompt-Injection Defense

Dec 2, 2025342

Anthropic confirms Claude 4.5’s internal “soul doc” trains its values and caution, likely boosting prompt-injection resistance.

AI Safety Prompt Injection AI Ethics Model Fine-Tuning

Under the Hood

Sparse Memory Layers: Targeted Continual Learning Without Forgetting

Nov 3, 2025102

Use sparse memory layers and TF-IDF–guided slot updates to learn continually without forgetting.

Continual Learning AI Architecture Model Fine-Tuning Catastrophic Forgetting

Under the Hood

Accents in 3D: How a HuBERT Model Maps English Accent Clusters

Oct 15, 2025260

A HuBERT model’s 3D latent map of English accents clusters by geography and social history more than by language-family taxonomy, offering an exploratory—but not definitive—view of accent relationships.

Data Visualization Model Fine-Tuning Speech Processing Computational Linguistics

Programming

Tinker: A Managed, Low-Level Fine-Tuning API for Open-Weight LLMs

Oct 1, 2025152

Tinker is a managed, flexible fine-tuning API for open-weight LLMs—spanning small to massive models—with low-level control, an open-source cookbook, and private beta access starting now.

Model Fine-Tuning AI Infrastructure Reinforcement Learning Open Source