“RoPE Distinguishes Neither Positions Nor Tokens in Long Contexts” RoPE is the default way LLMs track position, but this paper proves it breaks down as context gets long. Attention becomes close to random at telling near from far, and can even assign the same score to different
AI
-
Automating Podcast Playback Speed with Claude
By
–
Writing a program with Claude that slows down podcasts with @pmarca to 0.75x speed when he is speaking but switches to 1.5x speed for other speakers just to balence everything out.
-
Using Gemini 3.5 and Antigravity to Auto-Train AlphaZero for Go
By
–
Pretty wild to be able to automatically train AlphaZero and serve a playable Go demo with just 2 prompts using Gemini 3.5 Flash and Antigravity 2.0 … https://t.co/rsz3jQKqtG
— Thang Luong (@lmthang) 20 mai 2026Pretty wild to be able to automatically train AlphaZero and serve a playable Go demo with just 2 prompts using Gemini 3.5 Flash and Antigravity 2.0 …
-
Two real complexity classes: tractable and intractable in AI
By
–
There are only two real complexity classes in computer science: tractable (computing a function's output from its inputs) and intractable (computing a function's inputs from its output). AI deals with the latter. The rest is details.
-
UniVidX: a unified model for multiple video-generation tasks
By
–
What if one AI could handle multiple video generation tasks without needing separate models for each? Researchers from HKUST, Stanford, Tsinghua, and other top labs present UniVidX. It uses three simple tricks: random condition masking to let the model learn any input-output
-
Analyst uses ChatGPT to write report
By
–
Barclays Analyst Ajay Rajadhyaksha: ChatGPT, write my analyst report. Make no mistakes.
-
Gemini 3.5 Flash Performance Benchmarks and Capability Overview
By
–
Gemini 3.5 Flash is amazing! – Performs better than 3.1 Pro on coding & agentic tasks
– 4x faster than other frontier models
– 12x faster in @antigravity – 800 tokens/sec!
– Often at less than half the cost And Pro to come… Try it in @antigravity
, @GeminiApp & more – enjoy! -
Building Complex Multi-Agent Systems for App Development and Automation
By
–
🚨 Agent Swarms – Multi-Agents Delegate Complex Prompts To Sub-Agents
— Abacus.AI (@abacusai) 20 mai 2026
Use Gemini 3.5 Flash, Opus 4.7 and GPT 5.5 xHIgh to create complex multi-agent system
A master agent can orchestrate several worker agents to just do things
Build full-stack apps, mobile apps and automate… pic.twitter.com/5IDAvesIgwAgent Swarms – Multi-Agents Delegate Complex Prompts To Sub-Agents Use Gemini 3.5 Flash, Opus 4.7 and GPT 5.5 xHIgh to create complex multi-agent system A master agent can orchestrate several worker agents to just do things Build full-stack apps, mobile apps and automate
-
Technical evaluation of recurrent language models and evolution strategies
By
–
It actually holds up on real reasoning tasks like GSM8K and Countdown (not toy stuff) and matches GRPO pretty well. They even did full pretraining of a recurrent LM from scratch in pure int8. Sample efficiency is lower than backprop (normal for evolution strategies), but you get