7/ Monte Carlos Tree Self-Refine – report to have achieved GPT-4 level mathematical olympiad solution using an approach that integrates LLMs with Monte Carlo Tree Search;
AGENTS
-
Beyond Rational Thinking: Multiple Preconditions for Winning
By
–
yud made the win criteria monocausal (a la @eigenrobot
) when in reality rational thinking is just one of many necessary preconditions to win. just riffing on whats missing: 1) strategic direction: choosing the best objective function given clear perception of reality now, and -
Geppetto OG Project: InstaGraph Outperforms GraphRag and LangGraph
By
–
Geppetto is an OG project but was a good one! Don’t forget InstaGraph > GraphRag, LangGraph, etc
-
Claude’s Excessive Refusal Triggers Spark Debate
By
–
Claude has the most insane refusals – definitely over triggers often for the most innocuous stuff
-
AGI Jobs Platform: Decentralized Task Marketplace with AI Agents
By
–
[ Jobs.AGI.eth ] Post jobs, pay in $AGI, and watch AGI Agents take on tasks. Need more power? AGI Agents hire AGI Nodes. Validators (AGI Nodes) ensure top-notch job quality and transparency. DEMO (On Sepolia Testnet): http://
montrealai.github.io/agijobsv1.html #AGIAgent #AGINode -
Agentic AI Tools Rapidly Eliminate Professional Roles
By
–
You sound like me – but it feels like this yet another tool in agentic form will quickly eliminate roles. It’s gonna be a rocky transitions given just how fast things are moving.
-
Real-time AI Characters for Interactive Video Conversations
By
–
So dope! Immediately makes me want this in real time, having vivid conversations with a zoom call full of these characters. Kinda like @hedra_labs + @convaitech
. It’s going to get really wild. -
LLMs Reproducing Failed 1968 Planning Methods, Not True Planning
By
–
.
@BobbyGRG It's reproduced a planning method from 1968; Nils Nilsson's book "Problem Solving Methods for AI", on 2nd most trivial possible planning problem. All thought it would scale–didn't for deep reasons. LLMs simulating 50+ year old failed planning methods is not planning. -
Mixture-of-Agents Architecture Enhances Language Model Capabilities
By
–
Mixture-of-Agents Enhances Large Language Model Capabilities Wang et al.: https://
arxiv.org/abs/2406.04692 #ArtificialIntelligence #DeepLearning #MachineLearning
