SIGNAL
Tracking the global AI frontier — labs · research · agents · policy
Frontier Signal
⚡

Briefing

The fast feed: breaking AI news from the global tech press, deduplicated and time-ordered.

The fast feed: breaking AI news from the global tech press, deduplicated and time-ordered.

Latest in Briefing

30 stories
Beyond Visual Forensics: Auditing Multimodal Robustness for Synthetic Medical Image Detection
Research

Beyond Visual Forensics: Auditing Multimodal Robustness for Synthetic Medical Image Detection

With the rapid adoption of generative AI, synthetic medical images pose growing risks, including diagnostic deception and insurance fraud. Although prior work h…
Story Operators: Decomposing the Original $\to$ Sequel Transformation in Embedding Space
Research

Story Operators: Decomposing the Original $\to$ Sequel Transformation in Embedding Space

I treat a book as a point in a sentence-embedding space and a literary transformation as an operation on points. Given an original novel and its sequel, I ask w…
BrainAgent: A Large Language Model-Driven Multi-Agent Framework for Autonomous Brain Signal Understanding
Research

BrainAgent: A Large Language Model-Driven Multi-Agent Framework for Autonomous Brain Signal Understanding

Brain-Computer Interfaces (BCIs) and brain signal understanding are pivotal for clinical health and next-generation interactions. Despite this significance, its…
\chisao{}: A GPU-Native Parallel Optimizer for Multimodal Black-Box Functions via Convergence-Anticonvergence Oscillation
Research

\chisao{}: A GPU-Native Parallel Optimizer for Multimodal Black-Box Functions via Convergence-Anticonvergence Oscillation

Finding all modes of a multimodal black-box function is a fundamental challenge in optimization, Bayesian inference, and scientific computing. Existing approach…
SidConArena: An Environment Evaluating Agents in Open-Ended,Positive-Sum Bargaining Game
Research

SidConArena: An Environment Evaluating Agents in Open-Ended,Positive-Sum Bargaining Game

Evaluating LLM agents requires dynamic environments that go beyond static reasoning and zero-sum games. Real-world economic interaction is often open-ended and …
SurgAtlas: A Large-Scale Surgical Video-Language Dataset with 2,391 Hours of Open and Minimally Invasive Surgery
Research

SurgAtlas: A Large-Scale Surgical Video-Language Dataset with 2,391 Hours of Open and Minimally Invasive Surgery

We introduce SurgAtlas, the largest surgical video-language dataset to date, comprising 15,291 videos (2,391 hours) spanning 18 surgical specialties and over 5,…
Variable Bound Tightening for Nash Equilibrium Computation in Multiplayer Imperfect-Information Games
Research

Variable Bound Tightening for Nash Equilibrium Computation in Multiplayer Imperfect-Information Games

There has been significant recent progress in algorithms for approximation of Nash equilibrium in large two-player zero-sum imperfect-information games and exac…
Phonetic and semantic analyses of spoken corpora of Beijing and Taiwan Mandarin indicate that the neutral tone is a lexical tone
Research

Phonetic and semantic analyses of spoken corpora of Beijing and Taiwan Mandarin indicate that the neutral tone is a lexical tone

The neutral, or floating, tone of Mandarin Chinese is a tone with an enigmatic set of properties. It has been described as a reduced tone, or as a tone that som…
Geometry-Aware MCTS for Extremal Problems in Combinatorial Geometry
Research

Geometry-Aware MCTS for Extremal Problems in Combinatorial Geometry

We study certain extremal problems in combinatorial geometry that ask about configurations of points in an $n \times n$ grid that satisfy strict, global geometr…
A probabilistic framework for online test-time adaptation
Research

A probabilistic framework for online test-time adaptation

This paper presents a probabilistic framework for online test-time adaptation problems. In them, a model is trained on labeled data but must adapt to unlabeled …
Mean-Field PhiBE: Continuous-Time Mean-Field Reinforcement Learning from Discrete-Time Data
Research

Mean-Field PhiBE: Continuous-Time Mean-Field Reinforcement Learning from Discrete-Time Data

This paper addresses model-free continuous-time mean-field control in a setting where the population dynamics evolve continuously according to an unknown McKean…
SocialPersona: Benchmarking Personalized Profiling and Response with Multimodal Social-Media Context
Research

SocialPersona: Benchmarking Personalized Profiling and Response with Multimodal Social-Media Context

Personalized language-model assistants are often evaluated through a memory lens: can a model recall preferences users have explicitly stated in dialogue? More …
Scientific discovery as meta-optimization: a combinatorial optimization case study
Research

Scientific discovery as meta-optimization: a combinatorial optimization case study

Scientific discovery is fundamentally an optimization problem, defined by a vast "state space" of theories and experiments, and an evaluation criterion based on…
Reasoning Quality Emerges Early: Data Curation for Reasoning Models
Research

Reasoning Quality Emerges Early: Data Curation for Reasoning Models

Supervised fine-tuning (SFT) on a small, high-quality set of long reasoning traces is an effective approach for eliciting strong reasoning capabilities in Large…
SatSplatDiff: Geometry-preserving generative refinement for high-fidelity satellite Gaussian Splatting
Research

SatSplatDiff: Geometry-preserving generative refinement for high-fidelity satellite Gaussian Splatting

Gaussian Splatting has been recently explored for satellite 3D reconstruction, demonstrating flexibility and efficiency in representing radiometrically diverse …
Training Observable Control Policies to Expose Agent State Through Actions
Research

Training Observable Control Policies to Expose Agent State Through Actions

Physical or operational constraints often impose communications limitations on autonomous agents. Such limitations complicate monitoring or multiagent coordinat…
From Tokens to States: LLMs as a Special Case of World Models and the Continuous Path Beyond
Research

From Tokens to States: LLMs as a Special Case of World Models and the Continuous Path Beyond

The AI community has framed the relationship between large language models (LLMs) and world models as a dichotomy: LLMs predict tokens; world models simulate re…
Artifacts 22: Zyphra, Cohere, and Poolside are expanding the breadth of the ecosystem
Research

Artifacts 22: Zyphra, Cohere, and Poolside are expanding the breadth of the ecosystem

An assessment of the open ecosystem and the motivations behind releasing models…
HP Inc. launches Frontier strategic partnership with OpenAI
Labs

HP Inc. launches Frontier strategic partnership with OpenAI

HP Inc. scales its OpenAI Frontier partnership to deploy AI across customer experiences, software development, and enterprise operations.…
Only three AI models finished above starting capital in a 500-day startup survival test
Briefing

Only three AI models finished above starting capital in a 500-day startup survival test

Researchers at Princeton University built CEO-Bench, a test where AI agents have to run a fictional software company for 500 simulated days. Most current models…
Coinbase joins the rush to Chinese AI models as Western labs face a pricing stress test
Briefing

Coinbase joins the rush to Chinese AI models as Western labs face a pricing stress test

Coinbase CEO Brian Armstrong is switching his company to Chinese AI models like GLM 5.2 and Kimi 2.7. An automated routing system picks the best model for each …
AI won't become a real coworker until it stops answering and starts finishing tasks
Briefing

AI won't become a real coworker until it stops answering and starts finishing tasks

A survey paper by Tencent and several Chinese universities traces the path from chatbot to "digital colleague." AI systems won't become reliable coworkers, the …
Chinese cybersecurity firm builds AI tools to rival Mythos and frames the race as cyber-nuclear deterrence
Briefing

Chinese cybersecurity firm builds AI tools to rival Mythos and frames the race as cyber-nuclear deterrence

360 founder Zhou Hongyi presents two AI security tools designed to compete with Anthropic's Mythos. One has already flagged 3,432 vulnerabilities. Zhou admits C…
Sina's open model VibeThinker-3B aims to show reasoning compresses well but factual knowledge doesn't
Briefing

Sina's open model VibeThinker-3B aims to show reasoning compresses well but factual knowledge doesn't

Sina Weibo's VibeThinker-3B has just three billion parameters but matches models like DeepSeek V3.2 and Kimi K2.5 on math and coding benchmarks. Those models ar…
Large Language Model-Assisted Cleaning of Report-Derived Labels in a Large-Scale Chest CT Dataset
Research

Large Language Model-Assisted Cleaning of Report-Derived Labels in a Large-Scale Chest CT Dataset

Purpose: To evaluate whether large language model (LLM)-assisted label cleaning can identify label-report discordance in CT-RATE, a large-scale public chest CT …
Not All Claims Are Equally Risky: FACTOR for Adaptive Verification in Factual Long-Form Generation
Research

Not All Claims Are Equally Risky: FACTOR for Adaptive Verification in Factual Long-Form Generation

Large Language Models (LLMs) generate fluent long-form text, however, often add unsupported factual claims. Existing verification techniques improve factuality …
ROMEVA: Geometry-Preserving Vocabulary Expansion for Roman Urdu Language Models
Research

ROMEVA: Geometry-Preserving Vocabulary Expansion for Roman Urdu Language Models

Multilingual Language Models like mBERT are widely used for low-resource NLP, yet their adaptation to morphologically inconsistent languages such as Roman Urdu …
FetSelect: Task-Specific Architectures and Self-Supervised Learning for Automated Fetal Ultrasound Frame Selection
Research

FetSelect: Task-Specific Architectures and Self-Supervised Learning for Automated Fetal Ultrasound Frame Selection

Automated frame selection for fetal biometry remains under addressed, with most prior work targeting generic quality assessment or downstream measurement pipeli…
Generative Relightable Avatars
Research

Generative Relightable Avatars

We present Generative Relightable Avatars (GRA), a person-specific method for photorealistic free-view rendering and environment-map relighting of full-body hum…
Joint Air Traffic Flow and Capacity Management via Answer Set Programming
Research

Joint Air Traffic Flow and Capacity Management via Answer Set Programming

Operational Air Traffic Flow and Capacity Management (ATFCM) balances flight demand with available sector capacity, to ensure safe and efficient operations. Mat…