Thinking of ACE? We Can Do It with Fewer Tokens
Reducing computational overhead in AI systems could make advanced reasoning accessible to smaller models and lower-cost deployments.
Get smart on it
Two recent systems enable AI agents to learn from their own past failures without retraining. Both refuse to compress lessons learned into simplified summaries, but they differ in how they deliver those lessons to the model at inference time. One system injects a comprehensive playbook on every step regardless of the model's capacity, while the other tailors delivery to match what each specific model can actually use. Testing shows that selective delivery achieves equal or better accuracy while using a fraction of the tokens required by comprehensive injection.
Related news
Fine-Tuning Tool-Calling LLMs: A Complete Guide Using XYZ-Aquila-SFT and Qwen3
In this tutorial, we implement an end-to-end supervised fine-tuning pipeline for the XYZ-Aquila-SFT dataset, Hugging Face Transformers, PyTorch, and PEFT. We stream and inspect the dataset, parse multi-turn tool-use trajectories, extract structured tool calls, analyze corpus characteristics, and preserve embedded reasoning and observation patterns. We then convert tool schemas between message-embedded and structured formats, render Qwen-compatible ChatML with assistant-only loss masking, prepar
Meta’s ‘open’ AI, and a $250M deal gone very wrong
Meta released Glimmer this week, an open-weight AI model anyone can download and run on their own hardware — a contrast to Muse Spark, the company’s more powerful model that stays locked behind its own APIs. The release landed alongside a letter from Mark Zuckerberg arguing AI should be “for everyone” rather than controlled by a handful of labs, but as Equity’s […]
