We present FlashRender, a few-step generative rendering framework that retakes a source video along a target camera trajectory in seconds. We identify sampling-step-dependent camera control as a prominent manifestation of discretization error in existing multi-step generative rendering models and show that resolving this inconsistency substantially lowers denoising trajectory curvature, facilitati
In low-resource settings, deploying TTS typically requires choosing between a large voice-cloning model with costly inference or a compact fixed-voice system that requires a speaker-specific corpus. We study a third route: using a large voice-cloning model as a programmable data source to turn a short voice reference (e.g., 15 seconds) into a compact fixed-voice student trained entirely on synthet
Large language models achieve superior performance on tasks that require extended reasoning, but long chains of thought make the KV cache a severe memory bottleneck. Existing KV cache compression methods share one paradigm: score each cached token by some estimate of how much it will matter later, and keep the top-scoring ones. We show that the selection signal contributes almost nothing. Random A
Personalized assistants should not only comply with user requests but also assess whether those requests are appropriate given the user's current circumstances. However, prior work has primarily focused on accurately executing requests, overlooking the need for assistants to account for context and engage in conflict-based refusal. Furthermore, while existing work on conflict or safety detection r
Large Language Models (LLMs) often help users generate artifacts through iterative cycles of generation and revision in conversation. A challenge here is that, when users specify only a local change during revision, LLMs must instead identify the relevant dependencies and propagate the revision to all affected parts of the artifact. This paper studies this ability of LLMs on conversationally gener
A reasoning model can improve from its own on-policy experience, but this inner loop is fragile: terminal verifiers provide reliable yet sparse supervision, while dense same-model guidance can reinforce false confidence or overconcentrate learning on a narrow solution mode. We introduce FlowBalance, a verifier-grounded self-improvement method that learns a normalized distribution over complete res
The ambition of the 2025 PNPL competition (Landau et al., 2025) was to launch a multi-year curriculum for non-invasive speech decoding. Designed to progress from foundational tasks toward the linguistic complexity required for a practical brain-computer interface (BCI), it set the stage with speech detection and phoneme classification tasks. Winning submissions reached F1-macro scores of 95.6% and
497 points, 324 comments
Learn how to build a generative AI-based support operations platform on AWS that converts training videos into structured SOPs, applies Retrieval-Augmented Generation to guide ticket resolution, and uses machine learning to predict SLA risk and prioritize work.
Business intelligence dashboards can fail silently, showing blank, stale, or wrong data even when every infrastructure monitor reports healthy. Learn how an AWS team built an automated, AI-powered content validation solution on Amazon Bedrock that scans hundreds of dashboards and alerts owners, cutting mean time to detection from days to under an hour.
Learn how a global interdealer broker built an automated architecture documentation pipeline on Amazon Bedrock AgentCore that analyzes .NET code bases, generates architecture diagrams, and maintains searchable documentation through Amazon Bedrock Knowledge Bases and AWS CodePipeline.
Learn how University Startups and its AWS partner g/d/n/a scaled Trinity, a conversational AI solution for students with disabilities, into a serverless multi-agent architecture on Amazon Bedrock that produces IDEA-aligned transition plans for school districts across the US.
AI models have clearly proven their ability to discover and exploit vulnerabilities without much, if any, human assistance. To help defenders gain the advantage with AI, we built the Mantis harness to automate the discovery, triage, reproduction, and patching of software vulnerabilities.
940 points, 532 comments
A new method, called CW-Net, translates the reasoning process of an autonomous vehicle’s AI system into understandable concepts that explain its behavior.
This paper documents the acoustic-radio fusion in Techtile (ARFT) dataset, a synchronized measurement campaign for distributed wireless sensing and positioning in the Techtile testbed. Ultrasonic and radio frequency (RF) signals are simultaneously transmitted and captured at multiple positions in a 2D spatial grid inside the Techtile testbed. Each acquisition cycle corresponds to one rover stop, o
143 points, 24 comments
359 points, 166 comments
411 points, 176 comments
ATV Big Air Tour uses ChatGPT Work to speed up marketing, merchandising, and more. It even turned merchandise photos into an inventory website in 15 minutes.
167 points, 106 comments
Light amplification via Raman scattering in plasma has been severely constrained by stringent phase matching conditions and the need for plasma uniformity. To overcome these limitations, we propose a forward Raman amplification scheme that employs a positively chirped seed pulse co-propagating with a pump pulse in a nonuniform plasma with an upramp density profile. We demonstrate that the phase de
We’ve built an AI agent that acts as a secondary expert for a given domain, making deep specialist knowledge readily available and preserved for anyone in an organization to access, share, and build upon. This is not a typical domain-specific agent. Its novelty comes from integrating two layers: A structured, auditable knowledge architecture separates what [...] Read More... The post An Organizati
The rapid proliferation of large language models (LLMs) and the growing diversity of their applications presents a unique optimization opportunity: selecting the right model for the task, while optimizing for speed, cost, and quality at a per-task level. However, inference endpoints can vary widely in quality, price, latency, context support, tool use, domain expertise, and reasoning behavior. Thi
As ambitious space telescope concepts such as ultra-lightweight planar diffractive optical elements (DOEs) emerge, validating the performance remains a major computational challenge. Conventional Fourier propagation algorithms were observed to fail at meter-class apertures due to severe memory limits caused by rigid grid-sampling requirements, and the scaled-down proxy models used for reflector te
Two questions come up in almost every enterprise evaluation: where exactly does the inference run, and what do you keep? Data residency pins your Text-to-Speech, Speech-to-Text and Speech-to-Speech sessions to our EU servers, or to our US servers, enforced at the server and reported on every response, so you can assert on it instead of trusting a hostname. Zero Data Retention decides whether the t
Sep 2, 2026
The senior lecturer, already director of the degree program, will now oversee all aspects of the center’s activities and operations.
Learn how enterprise voice AI works and use seven call-readiness tests to evaluate hearing, latency, action, guardrails, and handoff.
Long-horizon manipulation is partially observable: the information needed to choose the next action may appear only in observations from minutes earlier. Existing memory mechanisms: retrieval banks, learned compressors, recurrent states must decide what to keep from the past before knowing what a future decision will require. This was motivated by the assumption that minute-scale history is too la
We study a governed approach to enterprise analytics: a language model interprets the question, while deterministic policy selects and runs a pre-approved analytical program that returns both results and evidence. We show that this restriction can remain expressive within a defined analytical class, using relational operations plus aggregation, comparison, windows, ranking, and similarity. Fixed m
Robot learning increasingly depends on broad and diverse demonstrations, yet collecting robot data remains expensive and poorly suited to covering the long tail of real-world tasks. To address this bottleneck, we introduce RoboTok, an internet-scale data engine that, given a query human manipulation video, retrieves manipulation-relevant human demonstrations from web videos for training dexterous
Visual fluency in generated video does not imply physical reliability, and a scalar quality score alone is incapable of indicating the obligation a clip violates or the moment it fails. We present VeriPhy, an auditable physical-verification system in which a text-only planner compiles the prompt into typed physical obligations and a statically validated execution plan before any frame is observed.
Many NLP tasks, such as summarization and extractive question answering, reduce to retrieving relevant content from documents under two constraints: coverage, retaining enough pertinent information to achieve some goal, and conciseness, removing as much irrelevant information as possible. Conformal prediction methods have been used to guarantee coverage, and must be optimized for conciseness throu
Causal inference is the practice of estimating the effect of a treatment or intervention from data. It traditionally requires a bespoke pipeline for every new problem: first proposing a causal mechanism, selecting a compatible estimator, and finally training it. Meanwhile, across diverse settings and modalities, much of machine learning has shifted to the paradigm of foundation models: networks pr
On-policy distillation (OPD) accelerates post-training by providing dense token-level supervision from a frozen teacher on the student's own rollouts. Vanilla OPD applies this supervision uniformly across prompts, without checking whether the teacher is reliable for each prompt. Because reverse KL is mode-seeking, a confidently wrong teacher can induce a strong yet misleading update. Distributiona
Speech brain-computer interfaces (speech BCIs) translate neural activity into language, offering a path towards restoring speech for people with paralysis and, more broadly, enabling new forms of natural human-computer interaction. Despite this promise, the field lacks a common measure of progress because systems use different datasets, recording methods, types of speech, and vocabularies, so thei
We introduce SolarWM, a fully open foundation for building interactive video world models from data preparation through long-horizon inference. Training across heterogeneous data sources and video backbones is challenging: datasets differ in temporal scale, camera geometry, visual quality, motion, and captioning styles, while video generators use distinct representations and architectures. Naive d
We introduce the Graph Machine (GM), an architecture that maintains an O(n)-sized state and accesses it through sparse, dynamic routing. Unlike methods with fixed-size states or sparse but static routing, GM preserves O(n) complexity in its sparse layers without restricting the potentially accessible state size to O(1). Instead, GM uses edges - pointer-like objects updated differentiably by a refe
Competitive programming has become a key test of large language model reasoning, with international competitions such as IOI and ICPC representing its most challenging settings. We present an end-to-end specialization pipeline combining large-scale problem curation, synthetic reasoning traces, supervised fine-tuning (SFT), and reinforcement learning (RL). Using 22,000 curated problems, we train Ne
Reinforcement learning with verifiable rewards (RLVR) has emerged as a powerful paradigm for large language model (LLM) post-training, but its reliance on coarse outcome rewards leads to limited guidance on intermediate reasoning processes. Existing approaches such as process reward modeling and on-policy distillation introduce additional constraints, such as reliance on a specialized reward model
Traditional speaker-attributed ASR systems treated ASR and speaker diarization as two separate tasks. Recently, end-to-end models such as VibeVoice-ASR have unified the two tasks within a single model. However, existing unified models still mainly support offline recognition, making it difficult to meet the low-latency requirements of real-time voice assistants and agents. To tackle this issue, we
Evaluating LLM agents is essential for guiding their development, yet it has grown prohibitively expensive: a single pass of a frontier model over an agentic benchmark can cost hundreds to thousands of dollars, a price paid repeatedly across iterative development cycles. Prior efforts, centered on benchmark distillation, reduce the number of evaluation tasks but leave the cost of executing each re
Streaming video understanding is a critical capability for real-world applications, including embodied intelligence, autonomous driving, industrial monitoring, surveillance and early warning, and wearable assistants. However, processing continuous video streams with multimodal large language models (MLLMs) is computationally expensive. Existing efforts have explored reducing streaming overhead thr
Training data attribution (TDA) aims to identify training examples that shape model behavior, but its intervention value depends on both which examples are selected and how they are modified. Influence functions (IF) estimate behavioral changes under infinitesimal reweighting, yet IF-selected examples often show limited advantages over random selection under conventional weight-based interventions
Multi-agent LLM systems commonly use an orchestrator to decompose a task for a team of workers and then improve through textual reflection. Despite strong empirical results, these systems lack a unified account of coordination, memory improvement, and the role of external verification. We model orchestrator-worker interaction as a bilevel coordination game: under bounded coupling, the workers' loc
Autonomous agents are beginning to carry out machine-learning (ML) research end to end. These agents combine a model backbone with a harness for planning, execution, memory, and verification, but this architecture still leaves domain-specific know-how outside the agent. We call this missing layer operational knowledge, the know-how that separates knowing a method from making it work. That knowledg
Language models spend most of their attention on a small fraction of context, yet they read the entire KV cache to find the few tokens that matter. If the user asks about a previous detail in a 1M-token conversation, global attention layers must scan the full context to generate each token of the reply. A prominent approach mitigates this cost by pre-selecting relevant tokens via lightweight proxy
Model compression techniques such as pruning and quantization facilitate the efficient deployment and acceleration of Large Language Models (LLMs). However, recent studies show that weight sparsification methods, such as SparseGPT, can amplify existing biases in models, with outputs varying significantly depending on persona cues in the prompt. In this paper, we introduce Debias-SparseGPT, a post-
We study the dynamics of Stochastic Gradient Descent (SGD), which is known to steer deep neural networks toward invariant sets that correspond to simpler subnetworks. How this steering unfolds over time remains poorly understood. We answer this by modeling the stochastic gradient flow (SGF) as a percolation process, in which architectural symmetries force subnetworks to merge in discrete simultane
Joint audio-video generation models have made substantial progress in visual quality and audio-visual synchronization. However, they still provide limited control over when shot transitions occur and dialogue is spoken. This limitation constrains their application in script-driven content creation, where timing errors can undermine narrative coherence and the viewing experience. Current joint gene
Faithfully translating research papers into repository-level implementations remains challenging because papers often describe methods at a high level, leave implementation assumptions implicit, and require generated repositories to preserve method logic, evaluation protocols, and cross-file consistency. Despite recent advances in paper-to-code agents, their intermediate outputs are often presente
Post-hoc calibration corrects reported confidence, yet a multiclass calibrator can also change the associated top-1 prediction. Accuracy captures only the net effect of these changes on correctness, not how often predictions change; the Top-1 Prediction Change Rate (TPCR) instead measures this frequency. We propose Calibrator-Output Repair for Top-1 Decision Preservation (CORD), the first post-fit
LLMs are increasingly deployed as orchestrators that coordinate specialized subagents to solve complex tasks through natural language. However, in many important domains like game playing and robotics, the strongest available agents are not language models. Integrating non-language agents with LLMs would require verbalization: compressing their rich continuous representations into sparse textual s
Benchmarks for LLM-generated GPU kernels decide correctness with a few random inputs and a loose floating-point tolerance, and their verdicts now feed leaderboards and reinforcement-learning rewards. Recent work agrees these checkers are weak and patches them by hand---extra input distributions, fuzzing recipes, tighter tolerances---with no way to measure whether any patch suffices. We introduce m
Vision-Language-Action (VLA) models demonstrate strong generalization in robotic manipulation and navigation, but existing fine-tuning methods provide limited safety guarantees. Current approaches primarily rely on Lagrangian optimization that enforces safety through soft penalties on expected cumulative cost, often resulting in residual constraint violations or overly conservative behavior. Moreo
Claude Fable 5.1 is now available on Amazon Bedrock and Claude Platform on AWS. This post covers Claude Fable 5.1's improvements, the Enterprise Frontier Safeguards for keeping your data in a cloud environment you control, and how to start building with the model on Amazon Bedrock.
251 points, 183 comments
Climate & Sustainability
580 points, 657 comments
191 points, 46 comments
Basis, Clay, and Exa Labs use AI agents to improve onboarding, account management, and developer integrations. See what enterprise leaders can apply.
When Atos set out to upskill 400 engineers in agentic AI, hands-on learning was the missing ingredient. Over three days, engineers built multi-agent systems on AWS through an AI League event. This post explains why Atos chose the format, what engineers built and learned, and what other enterprises should consider.
As generative AI adoption scales, cost governance becomes a top challenge. Learn how Jamf built real-time, per-user spend enforcement for Amazon Bedrock using IAM Customer Managed Policies, an Amazon Athena cost view, and a serverless AWS Lambda loop that applies tiered model limits in near-real-time without disrupting active sessions.
Amazon Quick proof-of-concept projects often stall when security teams review the production plan. This post walks through designing dashboards, Spaces, knowledge bases, agents, and Flows with security controls that hold as you scale: dataset shaping, agent isolation, document classification, and approval gates.
t54 built x402-secure, a trust layer on Amazon Bedrock AgentCore payments that scores every endpoint before an autonomous agent pays it. See how session budgets, credential isolation, and a deterministic trust gate have governed more than 20 million agent-initiated transactions with no human in the loop.
Learn how ZS built a security-hardened Amazon SageMaker platform that balances developer agility with healthcare-grade governance, serving 1,000+ daily active users across 200+ SageMaker domains.
Boomi Scribe is an AI-powered agent on AWS that automatically generates documentation for enterprise integration workflows. Learn how Boomi uses Amazon Bedrock, Amazon SageMaker AI, Amazon S3, Amazon DynamoDB, and AWS Lambda to parse integration DAGs, generate detailed documentation, and compare component versions at scale.
251 points, 353 comments
This paper addresses the joint reduction of the peak-to-average power ratio (PAPR) and out-of-band emissions (OOBE) in affine frequency division multiplexing (AFDM) systems by selecting the pre-chirp parameter c2. While existing approaches typically optimize either PAPR or OOBE independently, the proposed method jointly considers both metrics. To this end, a weighted cost function combining PAPR a
Astra is the first OpenAI model to meet the Critical cybersecurity capability threshold under the Preparedness Framework, with stronger safeguards for release.
ChatGPT can now connect to trusted healthcare data, helping clinicians securely access patient context, medical research, and more.
Acquisition strengthens TEKEVER's leadership in AI-enabled Autonomy through integration of Flowcopter's innovative hydraulic transmission technology Bristol, UK, 01/September/2026 - TEKEVER, Europe's leading provider of AI-driven autonomous systems, today announced the acquisition of Flowcopter, the Edinburgh-based engineering company behind an innovative hydraulic transmission technology for aero
The PhD student works to give young people an active role in shaping digital technologies that can support their own well-being.
Standard continuous gravitational-wave searches with pulsar timing arrays (PTAs) neglect cosmological redshift, restricting their applicability to the local Universe. We introduce a redshift-aware PTA framework and apply it to the Parkes PTA Data Release 3, deriving the first direct constraints on supermassive binary black holes (SMBBHs) at the cosmic dawn. Evaluating our limits across a broad red
See how Gilbert + Tobin combines CEO-led commitment, rigorous governance, and human accountability to scale ChatGPT Enterprise and Codex across the firm.
9551 articles sourced historically · 100 per page