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

Research

Curated daily papers, lab research blogs and national AI research programs across the US, China, EU, UK, Japan and beyond.

Curated daily papers, lab research blogs and national AI research programs across the US, China, EU, UK, Japan and beyond.

Latest in Research

30 stories
Lifelong In-Context Learning with Transformers Requires Parametric Forms of Attention
Research

Lifelong In-Context Learning with Transformers Requires Parametric Forms of Attention

Lifelong continual learning remains an obstacle on the path to human-like intelligence. Modern transformers show sparks of intelligence with in-context learning…

Efficient and Trainable Language Model Test-Time Scaling via Local Branch Routing
Research

Efficient and Trainable Language Model Test-Time Scaling via Local Branch Routing

Test-time scaling improves language-model reasoning, but existing approaches often face a difficult trade-off: long chain-of-thought sampling remains single-thr…

Three Buddhist Vocabularies: Computational Stylometry of the English Pali Canon across Sutta, Vinaya, and Abhidhamma
Research

Three Buddhist Vocabularies: Computational Stylometry of the English Pali Canon across Sutta, Vinaya, and Abhidhamma

We present a computational stylometric analysis of the Tipitaka across all three Pitakas in English translation, extending earlier work on the Sutta Pitaka alon…

From Sounds to Scenes: A Benchmark for Evaluating Context-Aware Auditory Scene Understanding in Large Audio Language Models
Research

From Sounds to Scenes: A Benchmark for Evaluating Context-Aware Auditory Scene Understanding in Large Audio Language Models

Recent Large Audio Language Models (LALMs) have achieved remarkable progress in audio perceptual tasks across individual acoustic layers, including speech, soun…

Long-Term Simulation Exposes Cognitive-Developmental Risks in AI Companions
Research

Long-Term Simulation Exposes Cognitive-Developmental Risks in AI Companions

AI companions powered by large language models increasingly interact with cognition-developing users, including children and adolescents, creating risks that ma…

Beyond Next-Observation Prediction: Agent-Authored World Modeling for Sequential Decision Making
Research

Beyond Next-Observation Prediction: Agent-Authored World Modeling for Sequential Decision Making

Recent studies on world modeling for Large Language Model (LLM) agents typically formulate the learning objective as next-observation prediction. However, this …

PRISM: Feed-Forward Single-Image 3D Reconstruction via Geometric Warp-Residual Modeling
Research

PRISM: Feed-Forward Single-Image 3D Reconstruction via Geometric Warp-Residual Modeling

Reconstructing 3D scenes from a single image is a fundamental challenge in computer vision, with broad applications in virtual reality, robotics, and content cr…

Brevity is the Soul of Inference Efficiency: Inducing Concision in VLMs via Data Curation
Research

Brevity is the Soul of Inference Efficiency: Inducing Concision in VLMs via Data Curation

Inference efficiency is typically pursued by shrinking the model: distillation, pruning, quantization, and sparse routing each lower per-token cost while treati…

Reclaim Evaluation: A Lossy Memory Is Worse Than an Empty One
Research

Reclaim Evaluation: A Lossy Memory Is Worse Than an Empty One

A language model's memory can be worse than having no memory at all. Give a model a memory that kept a wrong conclusion but dropped the work behind it, and it e…

How Reliable Is Your Jailbreak Judge? Calibration and Adversarial Robustness of Automated ASR Scoring
Research

How Reliable Is Your Jailbreak Judge? Calibration and Adversarial Robustness of Automated ASR Scoring

Almost every paper on LLM jailbreaks and prompt injection reports an attack-success rate (ASR), and that number is assigned not by people but by an automated ju…

Fault of Our Stars: Behavioral Drivers of Rating-Sentiment Incongruence
Research

Fault of Our Stars: Behavioral Drivers of Rating-Sentiment Incongruence

When people share experiences online, they often express thoughts in two ways: a star rating and a written review. In sentiment analysis, ratings are widely use…

Evaluating LLMs on Real-World Software Performance Optimization
Research

Evaluating LLMs on Real-World Software Performance Optimization

Software performance optimization is a notoriously complex and manual task. Despite the growing use of Large Language Models (LLMs) for code refinement, we stil…

Disease-Centric Vision-Language Pretraining with Hybrid Visual Encoding for 3D Computed Tomography
Research

Disease-Centric Vision-Language Pretraining with Hybrid Visual Encoding for 3D Computed Tomography

Vision-language pre-training (VLP) holds great promise for general-purpose medical AI by leveraging radiology reports as rich textual supervision, yet existing …

Energy-Efficient CNN Acceleration with MSDF Digit-Serial Arithmetic on FPGA
Research

Energy-Efficient CNN Acceleration with MSDF Digit-Serial Arithmetic on FPGA

This paper presents an energy-efficient hardware acceleration of the convolutional layers in the U-Net architecture for image segmentation, implemented on FPGA.…

VPA-Guard: Defending and Benchmarking Image-to-Video Generation Against Visual Prompt Attacks
Research

VPA-Guard: Defending and Benchmarking Image-to-Video Generation Against Visual Prompt Attacks

Recent advancements in Image-to-Video (I2V) generation have transformed input images from simple appearance references into interactive control interfaces where…

Low-Complexity Policy Tessellations in Structured Markov Decision Processes
Research

Low-Complexity Policy Tessellations in Structured Markov Decision Processes

We study optimal-policy geometry in structured Markov decision processes. While approximate dynamic programming and reinforcement learning typically approximate…

Constraint Tax in Open-Weight LLMs: An Empirical Study of Tool Calling Suppression Under Structured Output Constraints
Research

Constraint Tax in Open-Weight LLMs: An Empirical Study of Tool Calling Suppression Under Structured Output Constraints

Tool Calling and Structured Output are two core capabilities of modern Agent systems, yet their interaction under joint deployment conditions remains insufficie…

Expresso-AI: Explainable Video-Based Deep Learning Models for Depression Diagnosis
Research

Expresso-AI: Explainable Video-Based Deep Learning Models for Depression Diagnosis

Given the widespread prevalence of depression and its consequential impact on individuals and society, it is crucial to obtain objective measures for early diag…

BitNet Text Embeddings
Research

BitNet Text Embeddings

LLM-based text embedders have substantially improved retrieval and semantic representation quality, but their deployment remains costly: large backbone models s…

UniTeD: Unified Temporal Diffusion for Joint Perception and Planning in Autonomous Driving
Research

UniTeD: Unified Temporal Diffusion for Joint Perception and Planning in Autonomous Driving

Diffusion models have shown strong potential for multi-modal planning in end-to-end autonomous driving. However, most existing methods confine diffusion to the …

Gaussian Mean Field Variational Inference can Overestimate Predictive Variance
Research

Gaussian Mean Field Variational Inference can Overestimate Predictive Variance

Mean Field Variational Inference (MFVI) is widely understood to underestimate posterior variance. By analysing conjugate Bayesian Linear Regression (BLR), we sh…

OPERA: Aligning Open-Ended Reasoning via Objective Perplexity-based Reinforcement Learning
Research

OPERA: Aligning Open-Ended Reasoning via Objective Perplexity-based Reinforcement Learning

Reinforcement Learning (RL) has enabled LLMs to excel in objective reasoning tasks such as mathematics and code generation. However, applying RL to open-ended t…

MiniOpt: Reasoning to Model and Solve General Optimization Problems with Limited Resources
Research

MiniOpt: Reasoning to Model and Solve General Optimization Problems with Limited Resources

Achieving strong optimization generalization across diverse optimization problems while requiring limited training resources remains a challenging problem for o…

AI Snitches Get Glitches: Towards Evading Agentic Surveillance
Research

AI Snitches Get Glitches: Towards Evading Agentic Surveillance

To better assist users with completing challenging tasks, AI agents mediate communications, access data, and interact with different APIs. Many employers (and e…

Semantic Consistency Policy Optimization for Reinforcement Learning of LLM Agents
Research

Semantic Consistency Policy Optimization for Reinforcement Learning of LLM Agents

Group-based reinforcement learning effectively post-trains LLM agents for long-horizon, sparse-reward tasks by deriving step-level credit from trajectory outcom…

Enhancing Brain MRI Anomaly Detection and Reasoning with ROI Rethink and Synthetic Data
Research

Enhancing Brain MRI Anomaly Detection and Reasoning with ROI Rethink and Synthetic Data

Medical vision-language models typically generate diagnoses through single-pass inference without indicating which image regions support their conclusions. This…

Variational Autoencoder Layer
Research

Variational Autoencoder Layer

Variational Autoencoders (VAEs) belong to a family of autoencoders with probabilistic properties, making them well suited for generating data by producing a smo…

Explainable Control Framework (XCF) based on Fuzzy Model-Agnostic Explanation and LLM Agent-Supported Interface
Research

Explainable Control Framework (XCF) based on Fuzzy Model-Agnostic Explanation and LLM Agent-Supported Interface

Increasing demand for precise and reliable control in complex scenarios has led to the development of increasingly sophisticated controllers, including data-dri…

Helpful or Harmful? Evaluating LLM-Assisted Vulnerability Patching via a Human Study
Research

Helpful or Harmful? Evaluating LLM-Assisted Vulnerability Patching via a Human Study

Software vulnerability remediation is a cognitively demanding task that requires specialized security expertise often lacking in general developers. In the mean…

Tensorion: A Tensor-Aware Generalization of the Muon Optimizer
Research

Tensorion: A Tensor-Aware Generalization of the Muon Optimizer

Common first-order optimizers, such as Adam, implicitly treat each parameter block as an unstructured vector, which disregards the multilinear weight structure …