Linear message flows are too restrictive : LLM chat interfaces should have a button near each previous message that allows you to branch out from that message with a different follow-up question.
@aymericroucher
-
Celebrating shift from JSON tool calls to code blobs
By
–
Finally! Congrats, a step in a great direction!
We've been stuck in this local optimum of "writing actions as JSON blobs with individual tool calls in them" for much too long. Just let models write tool calls in code blobs! -
Rediscovery of CodeAct and smolagents a year later
By
–
I think they're finally going to rediscover CodeAct and smolagents, 1 year later!
-
Importance of In-Context Learning in AI Reasoning
By
–
Why would in-context learning be that important in AI? (i'm not talking about just "storing new info", which could be achieved for LLMs by enriching prompt, but about "updating reasoning processes") Like, for a human, I get it, we are limited by both lifespan and memory size
-
AI Improves Traffic Light Timing with Adaptive Control
By
–
Great unforeseen use case of AI:
Adaptive timing of red light, times better than pre-programmed times! (no more waiting for green at midnight in empty streets) -
Book Explains AI Progress for Wider Audience
By
–
Mom, I wrote a book! ⇒ "Ultra-Intelligence", the book to make more AI-pilled I often had this question from friends: "How far will AI go?" It's hard to know for a normal French person: most content on AI here is either math- or code-heavy, or just false. So I wrote this
-
Models Writing Tool Calls in Code Is More Expressive Than JSON
By
–
That's a good direction! Let models write their tool calls in code snippets, it's much more flexible & readable & expressive than the lame JSON blobs of usual tool calling!
-
PrediBench: AI Models Profiting on Polymarket Benchmark
By
–
We're thrilled to introduce PrediBench, our first production at @presage_labs
! PrediBench a live benchmark that answers the question "could an AI model earn money on Polymarket?" TL;DR: Some models like Grok-4 or GPT-5 do beat the crowd of human betters, and they turn a profit! -
Discussion on Challenges and Fundamentals of AI Agents at Hugging Face
By
–
Nice to see that progress goes on on agents at Hugging Face! Computer-use agents are notoriously hard to build, as seen by the complete lack of real-world applications of Operator / ChatGPT Agent.
Turns out the fundamentals matter:
– a simple, adjusted action space
– Reasoning https://
x.com/amir_mahla/sta
/amir_mahla/status/1970488574140407963
… -
Discussion on Agent Autonomy and Task Solving Time
By
–
Time-horizon autonomy definition does not take into account imo that task solving time for a human is non-linear.
> Really second that ! That can be a strong advantage of agents, especially if the task is parallelizable +1 also on the UX side : it's hard to get the thing right.