The Archive

RoBERTa AI-text detectors fail on formal human writing; 'humanized' machine text becomes more detectable via statistical complexity signals.

Marginal attention weights across 60+ LLMs exhibit emergent structure conserved as token-intrinsic properties independent of model architecture.

S2S-JEPA applies JEPA framework to subseasonal-to-seasonal weather forecasting, predicting only physically-grounded slow components beyond two-week horizon.

Formalizes actionable indeterminacy in LLM agents: when to act, seek clarification, or repair constraints in preference reasoning with solver-grounded benchmark.

Causal graph clustering preserves identifiability under certain conditions; nonidentifiability in clustered graphs does not guarantee nonidentifiability in original.

AdsCVR: first e-commerce cross-video reasoning benchmark (2,483 videos, 6,110 QA pairs) evaluates multimodal models on comparative product analysis.

Suno is branching out from the world of AI music, launching a new feature that generates spoken voices based on scripts or prompted descriptions. Speech is now available in public beta across Suno's web and mobile platforms, and allows you to simultaneously generate voiceovers and background music to accompany them. "Music will always be at the heart of Suno and what we build. At the same time, our vision has always extended to other forms of human expression," Suno chief product officer, Jack Brody, said in the announcement. "Today, we're expanding what's possible in Suno with Speech: the fi...

On an afternoon in Seoul in March 2016, I watched a program I helped build put a stone on the fifth line of a Go board in what looked like a gift to its human opponent. Move 37 in game two of the five-game match looked so absurd that some commentators thought it was a…

Latent Space reports Pi 1.0 and Pi Durable releases with TypeScript support; coverage of AIE NYC event.

Interview with MIT PhD Alex Zhang on RLM research, PhD trajectories, and future research directions in AI.

Chatham Financial deploys OpenAI's Codex and GPT-5.6 to automate trade validation, reducing processing time from 30 min to <4 min.

President Trump reportedly asked for Grok's opinion before invading Venezuela and capturing Nicolás Maduro.

The court acknowledges AI search comes with consequences, but it's not an antitrust issue.

Guided Vision is launching in Gemini Live on compatible Android devices today to use AI to give real-time audio descriptions of anything you point your phone's camera at. By sharing your camera in Gemini Live, you can have Google's AI help with things like reading small text, describing your surroundings, finding or identifying objects around you, and describing details on specific objects. Like the VoiceOver Live Recognition feature Apple has added to the iPhone and Vision Pro, it's designed for people who are blind, have low vision, or want assistance in specific situations. In addition to ...

OpenAI is rolling out new shopping features for ChatGPT that let users virtually try on clothing and accessories using their own photos and save products they like to a Favorites library.

Google launched its first advanced chip into orbit to pave the way for space data centers.

OpenAI has parted ways with three safety researchers after an internal investigation found they mishandled sensitive company information, report says.

AI agents are becoming a standard part of development workflows, but general-purpose agents weren't built with specialized infrastructure software such as... AI agents are becoming a standard part of development workflows, but general-purpose agents weren’t built with specialized infrastructure software such as NVIDIA DOCA in mind. Without domain-specific knowledge, agents may fall back on guesswork. This is an issue in infrastructure development because every correction cycle takes time away from deployment. DOCA is the unified software platform… Source

GALA distills 3D Gaussian avatar animation into linear blendshape combinations for real-time inference.

KaliBench evaluates LLMs on executable Kali Linux cybersecurity tool commands with verifiable rewards.

RPG enables autonomous robot skill improvement through offline dataset reconstruction and simulation practice.

NEPA improves diffusion transformers by predicting next embeddings as conditioning instead of static embeddings.

ScholarCatalyst benchmark measures AI ability to identify prior work that inspired new computer science research.

SILSA generates high-resolution 3D shapes using compact sliding-window slice latents instead of voxel tokens.

VISTA harnesses multimodal models for long-horizon interactive reasoning with lossless visual memory and active retrieval.

TACO optimizes LLM fine-tuning by reducing optimizer state memory via ternary column-wise updates.

FERPO improves continuous control RL by optimizing policies using critic values without critic gradient differentiation.

Cost-augmented Schrödinger bridges on graphs use Feynman-Kac tilts for exact solutions without learned control.

Hierarchical continuous diffusion language models improve parallel decoding by maintaining token dependencies through shared continuous state rather than independent marginal sampling.