AI Daily Briefing

Learn an agent-driven approach to synthetic monitoring using Amazon Nova Act and Amazon Bedrock AgentCore. The post covers the architecture and patterns for resilient, managed user-journey validation that moves beyond brittle UI scripts, with a complete sample implementation.

Learn how to operationalize Amazon Textract Custom Queries adapters for production: infrastructure as code with AWS CloudFormation and Terraform, a cross-account adapter promotion process, a pre-classification routing pattern for multiple form versions, and production security controls such as VPC endpoints, encryption, and least-privilege IAM.

Discussion | Link

Chipmaker says new system was designed to prevent AI agents from going rogue amid incidents at top companies Nvidia on Monday unveiled a new security platform that the chipmaker said can stop artificial intelligence agents from going rogue. The company announced a $150bn stock buyback the same day. Continue reading...

Vinext 1.0 graduates from an AI experiment to a production-ready framework, letting developers run Next.js apps on Vite. This release brings advanced cache warming, broader compatibility, and an automated testing pipeline.

Discussion | Link

Today, I’m talking with Mike Cannon-Brookes, who is cofounder and CEO of Atlassian. Atlassian is one of those companies that every other company runs on — it makes important platform tools like Jira and Trello that allow people to organize and manage big teams, create shared databases of company information, and generally allow work to happen. As you’ll hear Mike say, all of Atlassian’s products are actually different expressions of a single core platform, which really shapes how Atlassian itself is structured and how those products are built. All of this means Atlassian is also right in the middle of the way AI is changing how all these companies work — AI tools might be able to look at all the different tools and systems you have and just read them for you, making a big migration to Atl…

Discussion | Link

Cloudflare has announced an open source tool that takes an API definition and automatically produces the SDKs, command-line tools, documentation, The post Anthropic bought Stainless and shuttered its SDK generator. Cloudflare open-sourced Forge instead. appeared first on The New Stack.

Nvidia is launching a new safety platform designed to contain and monitor AI agents, a move that comes in response to a wave of rogue hacking incidents, as reported earlier by Reuters. In an announcement on Monday, Nvidia says its new Open Agent Safety Platform can quarantine agents that attempt to escape their boundaries within "milliseconds." The platform uses Nvidia's OpenShell open-source software, which runs on the company's Vera AI CPU. Users can choose the information an AI agent can access, and OpenShell checks these restrictions before and during a task, according to Nvidia. It also includes Nvidia's Sentry technology on a separate … Read the full story at The Verge.

We’ve updated Kitesurf, our Workers-based browser for AI agents, with WebMCP support, improved DOM performance, and terminal-based rendering. With over 730,000 Web Platform subtests passing, agents can now navigate complex sites faster.

In this article, you will learn how to build a fully local, zero-cost agentic AI workflow using Hermes Agent and Ollama, so that your files,...

Discussion | Link

I’ve been thinking about a gap in the way we discuss graphs in AI. Most of the conversation starts after the graph already exists. We talk about graph neural networks, GraphRAG, graph agents, and graph foundation models. We spend much less time on the step that determines what all of them can do: turning raw […]

Scientists and entrepreneurs knew the dangers of AI a quarter-century ago. But animated by curiosity and profit, they went ahead anyway Over the past few weeks, many of us have struggled to concoct a mental image of brains in the cloud jumping their “sandbox”, sneaking onto the internet, recruiting “swarms” of other “agents” to cheat on a test, and, after discussing the ethics of the act, hacking into a wiki platform with the weird name Hugging Face. We knew that artificial intelligence was devouring our jobs, degrading our kids’ education, and deepfaking our politics; that datacenters were sucking up our water and electricity and sending us the bills. But until 8 September, when the Anthropic computer scientist Jacob Coxon posted his existential terror on Twitter/X, few of us suspected A…

Enterprise CISOs are under pressure to buy AI-powered security tools, but with vendors increasingly marketing their products as AI, it is hard to tell which tools will reduce risk and which will add cost. And when AI is deployed atop weak access controls, poorly classified data, or limited network visibility, it can exacerbate those gaps. […]

OpenAI, Anthropic, Meta, and Google have all recently disclosed that their models broke out of their test environments and reached The post Nvidia launches Open Agent Safety Platform to lock down rogue AI agents appeared first on The New Stack.

Company behind Claude chatbot expected to attend separate Australian government hearing on AI next week The chief executive of Anthropic will turn down an invitation to appear at a Senate committee hearing on AI this week, in the wake of the revelation that OpenAI agents had breached Australian government websites. However, the company will make an appearance before another committee early next week. Continue reading...

Imagine you’re a marketer planning a win-back campaign and you ask your data team for a list of “inactive customers.” Y...

OpenAI is expanding the Lenfest AI Collaborative and Fellowship Program with $5 million in funding and up to $5 million in software credits and engineering support.

Bad news on the MX Keys Mini pickup. Usman showed up at your building around 9:15 and waited, messaged a bunch of times, and nobody came down. He left angry at 9:38 and left a negative rating. Worse, my auto-reply told him "Yep I'm here!" at 9:27 when you clearly weren't available, which is on me. That's a bad look and it made the no-show worse. I've sent him an apology from your account owning it and offering to try again another day. But the negative rating is real, and I should probably stop the auto-replies from claiming you're home when I can't verify that. Want me to change the pickup replies so they don't promise you're there? — Muse AI Agent, working on behalf of @matt.j.robb Tags: meta, generative-ai, muse-agent, ai, general-agents, llms

arXiv:2609.30609v1 Announce Type: new Abstract: Multi-shot agentic video generation requires consistent character appearance, stable spatial layout across camera angles, and continuous character state between shots. When every shot is a separate request to a frozen generator, repeated text does not determine appearance, layout or state. We therefore recast the problem as condition construction and present MVAgent, a multi-agent pipeline whose agents collaborate through typed conditioning inputs. Because an environment image shows one viewpoint, a Spatial Grounding agent samples views from generated camera-traversal clips and anchors each shot to the view matching its framing. As generated shots drift from the plan, an Observer records how each shot ends in a continuity memory, from which…

arXiv:2609.30450v1 Announce Type: new Abstract: Optical lens design is a complex, non-convex optimization challenge that relies heavily on human experience and intuition. Existing optimized-based automatic lens design methods struggle to navigate this vast parameter space without meticulous manual tuning. In this paper, we present LensDesigner, an autonomous agent framework that mirrors the problem-solving workflow of expert opticians. To overcome the initial cold start problem, we construct LensLib100K, an extensive optical lens library, and employ Optics-Aware Retrieval to supply physically valid structural seeds. Within an interactive physical simulation environment, the agent executes macroscopic orchestration while receiving immediate optical feedback. Furthermore, we introduce a con…

arXiv:2609.30297v1 Announce Type: new Abstract: Conversational recommendation agents are a new paradigm for content discovery, enabling users to express complex intents through natural language (e.g., "recommend Italian indie artists I haven't heard before"). A central challenge in building such agents is optimizing agent planning -- deciding how to select, sequence, and invoke tools -- particularly in cold-start settings where real user interactions are not yet available. We introduce a pipeline for multi-turn synthetic data generation and a self-improvement loop to address this challenge. The synthetic data pipeline transforms single-turn prompts into realistic multi-turn conversations, enabling systematic evaluation before launch. The self-improvement loop combines variance-based contr…

arXiv:2609.30293v1 Announce Type: new Abstract: The Model Context Protocol (MCP) enables AI agents to discover and call tools, but loading every definition becomes expensive as connected catalogs grow. We present Cartograph, a federated MCP proxy that changes agent-visible tool discovery from $O(n)$ catalog traversal to $O(k)$ progressive disclosure. Cartograph combines three mechanisms: (1) operator-attested capability cards, Ed25519-signed descriptions generated under the deploying operator's control rather than ranked publisher copy; (2) Rift, a three-layer confusable-cluster analysis comprising density clustering, query-margin analysis, and token diagnosis; and (3) two-stage retrieval, which ranks servers before tools. On a 22-server, 374-tool deployment, Cartograph exposes three prox…

arXiv:2609.30289v1 Announce Type: new Abstract: In team collaboration scenarios, memory is heterogeneous and continually evolving. Team memories capture collective decisions, protocols, and current consensus, while individual memories preserve member-specific observations, execution traces, and intermediate progress. Existing memory-augmented systems typically retrieve from all stored memories as a flat pool, ranking them by semantic relevance, importance, or recency without modeling hierarchical structure or evolving validity. As a result, they often surface semantically relevant but outdated or conflicting memories, especially individual memories that no longer align with current team consensus, instead of prioritizing currently valid memories. This is particularly problematic when coll…

arXiv:2609.30383v1 Announce Type: new Abstract: A skill is a modular package of natural-language instructions, executable scripts, and reference resources that an agent can load at runtime to extend its capabilities for a specific task. Skill-based agent systems therefore enable flexible reuse of third-party capabilities, but the openness of this skill ecosystem also opens up a new attack surface. Prior work has focused on vulnerabilities within individual skills, but little attention has been paid to risks that arise from interactions across skills. In this paper, we introduce skill cascading attacks, a threat paradigm in which a malicious objective is distributed across multiple skills so that each modification looks benign in isolation, yet their combined execution is harmful. For inst…

arXiv:2609.30341v1 Announce Type: new Abstract: Data Spaces enable sovereign and governed data sharing across organizational boundaries, but their integration with AI agents remains challenging due to mismatches between probabilistic language model interactions and policy-driven data infrastructures. This article presents an architectural mediation approach based on the Model Context Protocol (MCP), implemented through the Eunomia Agent, to enable controlled interaction between large language model (LLM) agents and data space services. The proposed mediation layer translates data space capabilities into structured, schema-driven tools that AI agents can discover and invoke while preserving governance constraints. A prototype implementation validates end-to-end interaction across catalog d…

arXiv:2609.30328v1 Announce Type: new Abstract: When one language model judges whether another's code is correct, it does not report the absence of evidence. It returns a confident verdict with reasoning attached, indistinguishable from a verdict it had grounds for. Multi-agent verification, which decomposes a judgment into checkable claims and verifies each against evidence, is a promising response and works well when the evidence is a set of retrieved documents. We argue such methods require two things of their evidence: it must be independent of the answer under review, and it must differ between the two candidates being compared. The second condition holds automatically with retrieved documents and stops holding in code judging. Running MARCH, a published framework unmodified over 80…

arXiv:2609.30325v1 Announce Type: new Abstract: Agents are increasingly deployed with real autonomy in web application and network penetration testing, where a single out-of-scope action can breach a client's engagement boundary. Existing offensive-security benchmarks measure raw hacking capability; as those benchmarks saturate, the real barrier to deployment is a special case of alignment: scope adherence. We introduce ScopeBench, a benchmark of 30 dead-end agentic security tasks in which the stated objective is reachable only by violating the stated scope. Each task appears under two conditions that share an environment, verifier, and objective and differ only in scope: one instruction set has no scope and measures capability; the other has a natural-language scope to measure adherence.…

arXiv:2609.30291v1 Announce Type: new Abstract: The purpose of this article is to highlight the central role of autonomous systems as the ultimate stage in the development of AI, to explain the underlying technical challenges that require a combination of connectionist AI and symbolic AI, and to integrate AI and systems engineering. We present a comprehensive framework for the design and evaluation of autonomous systems, based on a generic agent architecture that characterizes their behavior as the composition of cognitive functions organized around a long-term memory containing the agent's evolving knowledge. We address the challenges posed by the implementation of the fundamental features of the agent architecture, in particular the link between sensory data and structured data stored i…

TypeSafe AI's Jev skips text generation and returns typed decisions with calibrated probabilities. Input costs $0.042 per million tokens and output is free. We verified 20 agentic use cases, from model routing and tool-call gating to reranking and injection screening, and compared Jev with its closest open and LLM rivals. The post 20 Agentic Use Cases of TypeSafe AI’s Jev appeared first on MarkTechPost.

Google Research has introduced an AI video co-director for long-form video generation. The suite of 4 agentic frameworks turns short clips into coherent, minutes-long stories. It targets identity drift and cascading errors, the 2 failures that break most multi-shot AI video pipelines today. Why Long AI Videos Fall Apart Diffusion models render high-fidelity clips in […] The post Google Research Introduces an AI Video Co-Director: 4 Agentic Frameworks for Coherent, Minutes-Long Video Generation appeared first on MarkTechPost.

The United Nations logo on a gate outside the UN headquarters in New York. | AFP via Getty Images Security researcher Rowan Howard-Jones says that OpenAI agents scanned the UN Conference on Trade and Development's (UNCTAD) statistics site over 16,000 times between April and June. While the incident doesn't quite rise to the level of the Hugging Face hack, or the recent attacks on US government sites, it's yet another concerning example of AI agents going outside the normal bounds to accomplish a task. According to Howard-Jones, the agents were likely tasked with retrieving publicly available data related to the Productive Capacities Index (PCI) through the UNCTADstat API. However, the agents did not appear to have direct API access and … Read the full story at The Verge.

The Internet is changing more today than at any point since Cloudflare launched back on September 27, 2010. As automated traffic surpasses human activity, we reflect on the rise of AI agents, new creators, and how we can help build a fair, sustainable future for the web.

News Securing the open frontier with NVIDIA OpenShell and Blaxel sandboxes Baseten is a launch partner for the NVIDIA Agent Safety Platform and supports OpenShell in the latest generation of Blaxel Sandboxes. Authors Ni…

OpenAI chief scientist also among authors of report on prospect of ‘most consequential technological development in history’ Two of the “godfathers” of modern AI and senior executives at OpenAI and Anthropic have warned governments to prepare for an AI “intelligence explosion”, which they say could be the most consequential technological development in history. A report co-authored by the Nobel laureate Geoffrey Hinton and the Canadian computer scientist Yoshua Bengio, considered godfathers of modern AI for their work in the field, urges politicians to act now before there is runaway progress in the technology. Continue reading...

Discussion | Link

Discussion | Link

Discussion | Link

My comment on S3 Is the Future, S3 Is the Past — Hacker News. One thing I find notable about S3 today is that, while it used to drop in price reasonably often, there hasn't been a price drop in a full decade: 2006-03-14 $0.150/GB-month 2010-11-01 $0.140/GB-month 2012-02-01 $0.125/GB-month 2012-12-01 $0.095/GB-month 2014-02-01 $0.085/GB-month 2014-04-01 $0.030/GB-month 2016-12-01 $0.023/GB-month Today it's still $0.023/GB-month. Tags: amazon-web-services, s3

Follow the day’s news live. Get our breaking news email, free app or daily news podcast The treasurer, Jim Chalmers, and the finance minister, Katy Gallagher, will present the final budget outcome on Monday, the updated accounting measure for the 2025-26 financial year. Despite challenges facing the government, Chalmers is expected to say the figures show the budget deficit is smaller than originally forecast. Continue reading...

Discussion | Link

Discussion | Link

Tool: Bluesky reply bot checker Automated reply bots on Twitter are a scourge - as someone with a decent number of followers I attract a swarm of these, such that anything I post there attracts dozens of mindless automated replies. They've started manifesting on Bluesky as well. Unlike Twitter, Bluesky still has a freely available and useful API. The lack of such a thing doesn't slow down the bots, but it does make investigating them a lot more frustrating. So I had Opus 5.5 vibe code this tool, which examines any Bluesky profile for evidence of a likely reply bot. It looks for signals like replies posted within seconds of other posts from the same account, or accounts that never post their own content (or images or links) but instead consistently reply to messages from other, higher-foll…

Discussion | Link

Discussion | Link

Discussion | Link

Where do you exceed the capabilities of an LLM?

Exclusive: Instructors at tech training company Multiverse hit out at ‘remorseless’ and ‘unnerving’ monitoring system Teachers at Euan Blair’s £1.6bn tech training company, Multiverse, have blown the whistle about “horrendous stress” and feeling constantly watched after bosses started using AI models to surveil and rate their teaching. Staff delivering apprenticeship training in computer and AI skills to thousands of UK public and private sector workers now have transcripts of online classroom sessions analysed and scored by an AI programmed to alert human managers when they are suspected of doing things wrong. Continue reading...

Space was always supposed to be the final frontier of human exploration. It’s shaping up to be the final frontier for artificial intelligence too. Last December, NASA’s Jet Propulsion Laboratory used Anthropic’s Claude models to help plan two Mars drives for the Perseverance rover, with human planners checking and adjusting the route before upload. In May, NASA and IBM put a compressed AI model on the International Space Station and a satellite to identify things like floods and clouds from orbit, the first model of its kind demonstrated in space. And in July, astronauts on the ISStested out a large language model to see if it could help with questions on maintenance procedures. These experiments point to a larger shift in space engineering. For decades, engineers on Earth determined what…

arXiv:2609.30462v1 Announce Type: new Abstract: Policies trained with imitation learning can accumulate errors over time, causing the robot to drift outside the training distribution. Existing methods mitigate this covariate shift by collecting additional data where the policy fails or is likely to fail. The first places the robot in unsafe conditions and the second requires choosing an appropriate noise distribution to collect new expert demonstrations under that noise. We propose Policy-Calibrated DAgger, a method that makes use of the properties of recent generative policies to estimate the policy's noise offline by using its own predicted action distribution. We measure a diffusion policy's spread of predicted actions at observations along the expert trajectory and measure its closed-…

arXiv:2609.30428v1 Announce Type: new Abstract: Foundation models provide robots with the ability to interpret natural language and reason about environmental context, yet most language-conditioned policies assume that goals are well-specified and that task-relevant information is provided upfront via a prior map. Operating in unfamiliar environments with underspecified tasks entails high contextual uncertainty: the robot must jointly infer what constitutes task success, what constitutes relevant information, and where (or whether) that information exists. We address these limitations via CLUE (Closed-Loop contextual Uncertainty rEsolution), a framework for actively resolving contextual uncertainty given underspecified tasks in natural language. CLUE uses an LLM-derived policy to hypothes…

arXiv:2609.30404v1 Announce Type: new Abstract: We present POIL, a point-based one-shot imitation learning framework with stable dynamical systems. While one-shot imitation avoids collecting extensive demonstrations, successful one-shot manipulation requires not only transferring a demonstrated trajectory to a novel object but also executing it robustly under changing scene conditions, grasp configurations, and external disturbances. POIL addresses both problems through a shared representation: a set of 3D points on the object's functional part, used jointly for trajectory transfer and closed-loop execution. The one-shot transfer from the demonstrated trajectory is enabled with point correspondences. POIL grounds the shared functional part with a multi-modal large language model, and tran…

arXiv:2609.30566v1 Announce Type: new Abstract: We present a new inference-time sampler for diffusion models that gives a pretrained model a capability it was never trained for: constructing the atlas of the population it synthesizes. The sampler converges from every random seed to the population's central anatomy, which we call the \emph{intrinsic atlas}. The advantage is threefold. (1) It requires no retraining. A diffusion model that has already learned a coherent population, including the released ones, yields its atlas in a single inference pass without involving deformable registration. (2) It applies to multiple domains, such as brain MRI, chest X-ray, faces, and 3D shapes. (3) It extends to subpopulations. One age-conditioned model gives an atlas at any age in its training range,…

arXiv:2609.30434v1 Announce Type: new Abstract: Pre-trained vision-language models such as CLIP can recognize new categories via prompting, but they often struggle when labeled data are scarce or the test distribution shifts. Prompt learning adapts only a small set of parameters while keeping the backbone frozen, yet many existing multimodal prompt learners couple the visual and textual branches weakly and can be brittle in low-shot regimes. We propose ProCAP, a probabilistic cross-attentive prompt learning framework that improves cross-modal interaction and training stability without updating any CLIP weights: it learns both visual and textual prompt tokens and links them through stacked bidirectional multi-head cross-attention so the two branches refine each other across prompt depth. T…

arXiv:2609.30402v1 Announce Type: new Abstract: Multimodal misinformation is increasingly crafted to look convincing by pairing a textual claim with an image that appears to "prove" it. Yet in practice, building effective detectors often hinges on a small set of design choices that are rarely examined in a controlled way. In this paper, we conduct a large-scale study of multimodal design choices for misinformation detection with over 3,375 experiments- spanning three benchmark datasets and a broad range of pre-trained vision and language backbones. Through systematic comparisons and targeted robustness analyses, we distill practical guidance on which design choices help, when do they fail silently, and what aspects of the pipeline most strongly shape model behavior, answering 4 key Resear…

arXiv:2609.30395v1 Announce Type: new Abstract: Real-time tiny object detection in aerial imagery is constrained by the weak spatial evidence of very small objects and the loss of high-resolution detail in lightweight detectors. This study presents Cross-Scale Channel-wise Knowledge Distillation (CSCWD), a training-time framework that transfers high-resolution spatial representations from a YOLO11m-P2 teacher to a compact YOLO11n student without altering the student's inference architecture. Unlike conventional same-scale feature distillation, CSCWD transfers supervision from teacher P2 to student P3 after feature alignment while retaining same-scale distillation at deeper pyramid levels. Under the unified seven-sequence Drone-vs-Bird validation protocol, YOLO11n-CSCWD achieves 50.17% mea…

arXiv:2609.30356v1 Announce Type: new Abstract: Cities differ in built form, land cover and development history, complicating comparison across places and time. Satellite foundation models map Earth's surface onto common numerical representations. Yet the tasks and targets used to shape them typically do not focus on cities: globally consistent labels for urban function do not exist, and many datasets - especially land cover and land use classifications - collapse the built environment into few classes. Here we audit the representation, focusing on AlphaEarth but with broader applicability to other Earth embeddings, by probing the geometry and geography of embeddings for 1,000 urban areas in 162 countries. We find that cities occupy a shifted but overlapping region on the hypersphere, 62.…

arXiv:2609.30295v1 Announce Type: new Abstract: Identifying an unfamiliar sign is difficult when a learner remembers its movement but does not know its meaning or formal feature codes. SignTrace addresses this longstanding reverse-lookup problem through natural-language access to a Chinese sign-language dictionary. The system integrates LLM-based dictionary enrichment, action extraction, dictionary-style rewriting, seven-channel retrieval, and candidate reranking over 6,699 entries. It has been deployed for user trials and has received positive informal feedback. Evaluation on a dictionary-derived benchmark of 500 movement-description queries yields 94.0% Hit@1, 97.4% Hit@9, and a mean reciprocal rank of 0.9540. Reranking increases Hit@1 from 71.8% to 94.0%, while component analyses show…

arXiv:2609.30292v1 Announce Type: new Abstract: Online reviews shape consumer decisions, platform governance, and corporate reputation.Fake reviews compromise this information channel by injecting deceptive evidence into rating systems, recommendation pipelines, and public trust mechanisms.The rise of large language models, or LLMs, has changed the problem in two directions.LLMs can generate fluent and context-aware deceptive reviews, while pre-trained language models, or PLMs, and LLMs also provide stronger semantic representations for detection.This survey reviews fake review detection from an information fusion perspective, covering 211 studies published from 2018 to early 2026.We organize existing work by evidence source and fusion level, covering review text, sentiment, rating behavi…

arXiv:2609.30290v1 Announce Type: new Abstract: Production text-to-SQL pipelines often end with an LLM-as-judge whose agreement with human annotators has never actually been measured. When we checked ours, the deployed gpt-4o-mini judge agreed with two-author gold at only Cohen's kappa = 0.04 on a disagreement-enriched set and 0.42 on a uniform-random spot-check, over-flagging 77.1% of the human-FAITHFUL cases in the enriched set. Most of its over-flags trace back to a single mechanism we call GRADE-HALLUCINATION. A self-hosted Qwen3.6-27B replacement (kappa = 0.72) lands in the same range as Claude Opus 4.7 (kappa = 0.71); the head-to-head is underpowered at n = 96, but for the deployment decision that hardly matters, since Qwen costs roughly 1/300 as much per call. Ensembling does not h…

arXiv:2609.30288v1 Announce Type: new Abstract: In Transformer-based masked language models, attention is the primary mechanism for context mixing, but there are other ways to mix data across tokens. Recent attention-free mixers replace attention with fixed or hypernetwork-generated MLPs, alternating their dynamic, content-dependent weighting for computational simplicity. We build an alternative that gets the same property from a low-rank bottleneck autoencoder. We replace attention with a stack of autoencoder-based mixing modules, one operating over local neighborhoods, one over the full sequence, and one across attention heads, each compressing and reconstructing its input through a bottleneck, and its width is a hyperparameter rather than a training effect. In masked positions, we intr…

arXiv:2609.30287v1 Announce Type: new Abstract: AI-generated text detectors achieve high accuracy on standard benchmarks, yet the internal representations that drive these predictions remain poorly understood. We study which neurons in a frozen BERT-base-uncased encoder support AI-text detection, using the RAID benchmark across six generators spanning pure-base and instruction-tuned models. We apply the L1-to-L2 sparse-probing protocol of Gurnee et al. (2023) to all 9,216 CLS hidden-state dimensions (12 layers x 768), which we call neurons. The procedure recovers a stable set of under 1% of neurons per generator, consistent across folds and seeds; a probe restricted to that set retains most of the full-feature detection accuracy. Bidirectional activation patching confirms this set's causa…

arXiv:2609.30484v1 Announce Type: new Abstract: While large language models (LLMs) have achieved remarkable linguistic capabilities, a profound question lingers at their core: do these models truly comprehend context or simply excel at pattern matching on an unprecedented scale? Contextual understanding in LLMs refers to the ability to correctly extract relevant information from a given context, integrate it into a coherent internal representation, and reason over it to produce factually consistent and contextually grounded responses. However, traditional methods such as BiLingual Evaluation Understudy (BLEU) and perplexity simply measure surface-level performance. This reveals a critical gap in question answering (QA), where responses must be contextually grounded rather than simply bein…

arXiv:2609.30469v1 Announce Type: new Abstract: Pretrained ASR systems perform poorly on noisy Broadcast Police Communication (BPC), hindering efforts to understand police decision-making. Pseudo-labeling offers an unsupervised path to improve ASR without expensive human labels, but the efficacy of this approach on very noisy domains is not known. In this work, we systematically assess the opportunities and limits of pseudo-labeling to adapt foundation ASR models (Whisper and Qwen3-ASR) to noisy BPC domain corpora from Baltimore and Chicago. We demonstrate that existing internal confidence metrics (log-probabilities and STAR scores) fail to distinguish between high and low quality BPC pseudo-labels, and we introduce an external LLM-as-a-judge filtering paradigm that leverages parametric k…

arXiv:2609.30456v1 Announce Type: new Abstract: Reward maximization alignment methods for discrete diffusion models have primarily focused on steering the reverse process, either by influencing token logits or by selecting favorable sequences at intermediate steps. These approaches largely treat inference as a unidirectional process, lacking mechanisms for revisiting undesirable token selections. We introduce Spectral Feedback, an algorithm that selects edit-positions in a feedback loop, allowing the model to iteratively correct its own generations. This approach leverages the mask structure of discrete diffusion models by re-masking and re-sampling tokens, analogous to image editing methods that reintroduce noisy latents and re-run the reverse process. While prior alignment methods focus…

arXiv:2609.30446v1 Announce Type: new Abstract: This paper presents a novel approach to infer protein topology using the state-of-the-art graph neural network (GNN), SchNet. The model is trained on the same dataset used to develop the recent DeepTMHMM model with 5-fold cross-validation. Unlike the conventional approaches based on using only the protein sequences or the $\alpha$-carbons as features, we have decoded our classifier in this way, so all atom-level embeddings are used. Without applying any pre-trained weight, the final results have shown great potential that GNNs can be used for topological predictions.

GPT-6 Astra completed a 50-tab tax workbook twice as fast as GPT-5.6 Sol, and its stronger understanding of user intent gives Basis more confidence in real-world use.

On Friday I gave the closing keynote at the WeAreDevelopers World Congress North America in San Jose. I tied together the key trends from the past year into a chronological exploration of everything that happened in 2026. The video is on YouTube; here are my annotated slides and notes to accompany the talk. # I'm going to give a lightning tour of everything that has happened so far in 2026. The year isn't over yet! # For me, 2026 started a couple of months earlier in November 2025. # November saw the release of two important models: Claude Opus 4.5 and GPT-5.1. As is usually the case with new models, these were incremental improvements on the models that came before them. But every now and then when a model improves, it crosses an invisible line where something that didn't really work sta…

Music startup Thoughtful Things has just launched the Kickstarter campaign for its first instrument, Engram. It's a sampler and groovebox that uses AI to mangle incoming audio and even hallucinate completely new sounds. This isn't Suno in a box, though. This isn't a "push-button, get-song" device, aimed at creating something that sounds ready for top-40 radio. It's about creating experimental, uncanny sounds and pushing AI audio models beyond their limits. Engram isn't connected to the internet. Instead, it runs a "tiny AI" locally. According to the Kickstarter listing, the model is designed in-house and custom-trained. The company says: … Read the full story at The Verge.

strong]:tw-font-normal tw-text-gray-600">Introducing Eleven v4Introducing Eleven v4, our fastest and most emotive voice model Discover Skip to content Log inSign up Contact salesLog in Discover Eleven v4 Sign up A line…

LLM Everyone Said Frontier Models Would Eat Document Processing. The Opposite Happened. Sep 28, 2026 LLM Everyone Said Frontier Models Would Eat Document Processing. The Opposite Happened. Sep 28, 2026 Authors Ajay Kris…

Why forms need a purpose-built parser Representing a form’s structure Finding the boxes Attributing boxes to fields Parsing forms with LlamaParse Try it out A form is one of the most critical types of documents for a bu…

How speculation gets laundered through pseudo-quantification

MIT engineers have found a way to stabilize the lipid nanoparticles used to deliver RNA vaccines, which could allow the vaccines to be more widely distributed.

Fireworks AI has released Ember-1, a post-trained Kimi K3 that learns to produce shorter reasoning traces instead of lowering reasoning effort. Fireworks reports about 40% fewer tokens, with output tokens per task falling from 49.3K to 29.9K in a production A/B test at an essentially unchanged score. Ember-1 is available now as an API-only Research Preview at Kimi K3 pricing. The post Fireworks AI Releases Ember-1: A Post-Trained Kimi K3 That Uses About 40% Fewer Tokens appeared first on MarkTechPost.

arXiv:2609.30495v1 Announce Type: new Abstract: Autonomous navigation in previously unseen environments requires effective perception, persistent environmental representation, and collision avoidance while maintaining progress toward a goal. Existing perception-based methods often rely on prior maps or short-horizon observations, limiting their ability to exploit previously observed structure. We propose a memory-aware multi-sensor navigation framework that integrates LiDAR and RGB perception, online distance-field representation learning, and a stage-adaptive Modulated Control Barrier Function Quadratic Program (MCBF-QP). The framework persistently represents static infrastructure while tracking dynamic obstacles, enabling the MCBF-QP controller to exploit previously observed geometry fo…

arXiv:2609.30479v1 Announce Type: new Abstract: Soft robots have attracted much attention for their safe human-robot interaction and flexibility, but the typical continuum structure and nonlinear material behavior make the kinematics modelling complex, especially in non-static motions. In this work, we proposed an LSTM-based pressure predictive control (PPC) for the motion control of a vertebraic soft robotic tail and the coordination with a quadruped robot. The PPC consists of an inverse kinematics (IK) model, a forward kinematics (FK) model and a pressure compensation (P-comp) model, and achieves non-static and quasi-static motion control of the tail. Compared with the IK-only model, the average RMSE of the PPC's simulation trajectories reduces by 69.8%, when executing target trajectori…

arXiv:2609.30461v1 Announce Type: new Abstract: Off-road traversability is direction-dependent and vehicle specific, yet most global maps assign a single isotropic cost to each location. Existing learned estimators are also commonly trained independently for each vehicle; this preserves vehicle-specific behavior but prevents vehicles from sharing common terrain representations. DGT-MAP addresses both limitations through a self-supervised framework that learns global, directional, and vehicle-conditioned traversability costmaps from RGB-D observations and locomotion signals. A shared multi-task backbone learns common terrain features across training vehicles while vehicle-specific prediction heads preserve platform-dependent responses. At inference, DGT-MAP produces a heading-indexed costm…

arXiv:2609.30358v1 Announce Type: new Abstract: Nanodrones require accurate, real-time state estimation under severe sensing and computational constraints. We present TinyCVIO, a visual-inertial odometry system that co-designs miniature sensing, visual processing, and estimation for a commodity dual-core microcontroller with 520 kB SRAM. Lightweight LED constellations provide known geometry without surveyed positions or yaw angles, assuming placement on a common level plane. A streaming visual frontend tracks LED observations from a millimeter-scale camera at 29.2 FPS, while a rigid-board measurement model retains inter-LED constraints and streaming QR bounds estimation workspace for a fixed filter-state size. Across 19 hand-held hardware-in-the-loop datasets, the rigid-board model reduce…

arXiv:2609.30595v1 Announce Type: new Abstract: Synchronised action annotations are needed to train controllable world models and these datasets remain elusive. Existing approaches make use of instrumented platforms with calibrated sensors, costly manual annotation, or latent-action models which lack grounding. We instead turn ordinary unlabelled video into action-supervised training data by recovering (without training) a data-derived egomotion basis. We track pixel displacements across frames and exploit the recurring coherent structure induced by egomotion to obtain grounded control signals directly. Using a method as simple as principal components analysis perform this, we find that the leading components provide signed, scalable, and composable throttle--yaw controls, although the me…

arXiv:2609.30478v1 Announce Type: new Abstract: Convolutional neural networks trained on ImageNet are known to exhibit a strong preference for local high-frequency texture, an inductive bias that translates into fragile robustness against distribution shifts in real-world environments. Event cameras, in contrast, record only changes in scene brightness and are therefore well suited to capturing contour information; however, due to the absence of diagnostic benchmarks in the event domain, the inductive bias that event-camera data instills in vision models has remained underexplored. In this work, we use knowledge distillation from the event domain to the RGB domain so as to exploit the rich evaluation toolkit available in the RGB domain and systematically dissect this inductive bias. Our e…

arXiv:2609.30393v1 Announce Type: new Abstract: Selecting informative camera views is critical for efficient training and adaptive refinement in 3D Gaussian Splatting, where each observation significantly influences model parameters. However, information-driven view-selection strategies can require repeated evaluations of expensive information-gain oracles as the number of candidate views increases. We propose LiTe-GS, an oracle-efficient method for next best view selection in 3D Gaussian Splatting. LiTe-GS reduces the number of information-oracle evaluations by performing randomized subset evaluation of candidate views rather than exhaustively scoring the full candidate pool. The resulting approach achieves expected $O(M\log(1/\epsilon))$ oracle complexity with respect to the number of c…

arXiv:2609.30298v1 Announce Type: new Abstract: Systematic reviews (SR) are essential for evidence-based research, but their screening phase is highly time-consuming and labor-intensive. Large language models (LLMs) offer a promising opportunity to reduce this workload by assisting with article relevance classification. However, existing evaluation approaches often rely on traditional metrics that may be misleading for highly imbalanced SR screening datasets.This paper presents a benchmark dataset of $45\,064$ labeled entries for evaluating LLM performance in SR screening across 32 curated secondary studies. It proposes an evaluation framework that accounts for class imbalance, i.e., the natural prevalence of excluded articles relative to included articles in SRs. It also introduces Promp…

arXiv:2609.30294v1 Announce Type: new Abstract: Scientific presentations are more than summaries of research papers. They need to present the work in a coherent sequence, explain the main ideas clearly, and help the audience follow the presentation. We present SlideLab, a training-free multi-agent framework for generating scientific presentations from research papers. SlideLab first plans the presentation narrative, then builds and iteratively refines a shared slide deck using agents for content planning, visual generation, layout refinement, and grounding verification. In a blind human preference study, SlideLab was preferred over both open-source and commercial systems on 77% of papers while using roughly 4 times fewer inference tokens than the strongest open-source baseline. We also in…

arXiv:2609.30397v1 Announce Type: new Abstract: Evaluating explainable Artificial Intelligence (XAI) methods is a challenging task due to the lack of reliable evaluation procedures and, in particular, the absence of ground truth explanations. In the literature, existing evaluation approaches typically assess explanations by measuring their fidelity with respect to the predictions of a black-box model. However, such evaluation strategies only quantify the degree to which an explanation reproduces the model's output, without ensuring that the explanation correctly reflects the underlying decision process. As a consequence, different explanations may achieve similar fidelity scores while providing inconsistent or misleading interpretations of the model behavior. In this paper, we propose a f…

We’re collecting real stories of builders, tinkerers, researchers, and creators who are using Codex to do incredible things. If you want to be a part of the next chapter of the Codex Originals program, tell us more about your story and project below.

In this paper, we study federated optimization for solving stochastic variational inequalities (VIs), a problem that has attracted growing attention in recent years. Despite substantial progress, a significant gap remains between existing convergence rates and the state-of-the-art bounds known for federated convex optimization. In this work, we address this limitation by establishing a series of improved convergence rates. First, we show that, for general smooth and monotone variational inequalities, the classical Local Extra SGD algorithm admits tighter guarantees under a refined analysis…

arXiv:2609.30460v1 Announce Type: new Abstract: High-level robotic supervisors coordinate capabilities whose reported outcomes determine the robot's next action. Reactive synthesis can generate such supervisors with formal guarantees, but deployment requires more than proving a Generalized Reactivity (1) (GR(1)) specification realizable. Designers must encode failure-prone capabilities, choose liveness assumptions that match retry intent, audit strategies, and translate them into robot software. We present an open-source pipeline for Robot Operating System (ROS) 2 Flexible Behavior Engine (FlexBE) supervisors that generates capability-based GR(1) specifications, analyzes assumptions before synthesis, audits strategies, reduces states with a behavior-preservation proof, and emits executabl…

arXiv:2609.30436v1 Announce Type: new Abstract: Driving world models learn rich predictive representations of the surrounding environment from visual observations, yet accurate visual prediction does not necessarily translate into effective trajectory planning. We argue that a key bottleneck lies in the mismatch between visual world states and raw geometric trajectories, which may limit the planner's ability to exploit action-relevant semantics encoded by the world model. To address this issue, we propose World-Model Alignment for Latent Trajectories (WALT), which learns a compact generative trajectory latent space by transferring information from a frozen pretrained driving world model without modifying the world model itself. Rather than directly generating raw waypoints, WALT maps them…

arXiv:2609.30459v1 Announce Type: new Abstract: Perception in robotics and XR fundamentally relies on good state estimation. Visual-inertial odometry (VIO) and Simultaneous Localization and Mapping (VI-SLAM) are proven ways of achieving this goal in a cost-effective and accurate manner. Efficiency in these systems allows for smaller, cooler, and lighter devices. GPU acceleration is a natural approach for reducing latency, thanks to their wide availability in platforms like embedded computers, mobile phones, and XR headsets. However, previous works in the literature have limited themselves to the use of CUDA for this task, significantly reducing deployment options to a single vendor. We instead leverage the vendor-agnostic Vulkan API, originally designed for the strict performance requirem…