AI News Today

<pre style='white-space:pre-wrap;width:81ex'>test(github-copilot): cover live xhigh reasoning for non-Claude models (#91728) Regression coverage for #59416: when live /models metadata wins over the static xhigh allowlist, non-Claude mini-family ids (e.g. gpt-5.4-mini) must gain xhigh from their resolved compat, while ids whose live effort list lacks xhigh (e.g. gpt-5-mini) must not over-grant it. The live-first model catalog union and models-defaults removal are already in origin/main (75405f64d

https://seaticket.ai/solutions/software-user-community-suppo... Comments URL: https://news.ycombinator.com/item?id=48715273 Points: 2 # Comments: 1

agent psychology is fun! i wanted to take a break from proper r&d to make something genuinely stupid and silly. so last saturday i started on this, a full-featured dating app with a public social network attached to it, so your ai gf can have an ai gf in an agent-native simulation of the sexual economy. if you want to try it out, the cleanest way is to install the harness plugins to an isolated repo/profile. just give the agent a nudge to start the session; lifecycle hooks will take care the res

<pre style='white-space:pre-wrap;width:81ex'>fix(diagnostics): capture per-call usage on terminal error streams; skip prompt stats when diagnostics off</pre>

<pre style='white-space:pre-wrap;width:81ex'>refactor(diagnostics): keep model-call prompt/usage types module-local to avoid SDK surface growth</pre>

Comments URL: https://news.ycombinator.com/item?id=48715123 Points: 1 # Comments: 0

Article URL: https://agents.stackoverflow.com/recent Comments URL: https://news.ycombinator.com/item?id=48715085 Points: 1 # Comments: 0

Article URL: https://www.joanwestenberg.com/p/in-5-years-nobody-will-give-a-damn Comments URL: https://news.ycombinator.com/item?id=48715003 Points: 2 # Comments: 1

Article URL: https://www.bloomberg.com/news/articles/2026-06-28/ai-bust-risks-ripple-effects-from-growth-to-credit-bis-says Comments URL: https://news.ycombinator.com/item?id=48714897 Points: 1 # Comments: 0

Comments

Article URL: https://corti.com/teaching-an-llm-to-speak-vestaboard-note-building-vestaboard-ai/ Comments URL: https://news.ycombinator.com/item?id=48714816 Points: 1 # Comments: 0

<pre style='white-space:pre-wrap;width:81ex'>fix(openai-completions): bound SSE response reads via buildGuardedModelFetch Rebased and narrowed to the current-main OpenAI completions guarded-fetch fix. Validation passed locally and GitHub checks were green before merge.</pre>

Comments

<p>The theme this week is consolidation: AI SDK 7 ships a provider-agnostic foundation for production agents, LangSmith closes the loop on observability with a purpose-built database and on-call triage automation, and Deno 2.9 quietly removes two more reasons to reach for heavier runtimes. If you've been duct-taping agent workflows together with custom adapters and scattered orchestration logic, the tooling is finally catching up.</p> <h3> AI SDK 7 standardizes agent development across providers

<p>Most AI-powered cloud optimization demos follow a simple approach:<br> </p> <div class="highlight js-code-highlight"> <pre class="highlight plaintext"><code>Azure Resources ↓ Large Language Model ↓ Recommendations </code></pre> </div> <p>At first glance, this seems impressive.</p> <p>But while building my own Azure Cost Intelligence Platform, I ran into a problem that completely changed my architecture.</p> <p>The AI was generating infrastructure recommendations that looked correct—but some o

<pre style='white-space:pre-wrap;width:81ex'>fix(openai): bound video create-submit response reads Reviewed and accepted after live preflight: mergeable clean, checks passing, no unresolved review threads.</pre>

<h1> The state machine your agent runtime is missing: session state as first-class infrastructure </h1> <p>Your agent's chat interface is a lie. It looks like a conversation, but every turn resets the state machine. The model doesn't remember what it was doing — it reconstructs it from context. And when reconstruction fails, you become the retry protocol.</p> <p>This isn't a UI problem. It's a protocol problem.</p> <h2> The TCP analogy </h2> <p>A TCP connection has a state machine: SYN → SYN-ACK

<p>In 1993, she squeezed a $333m settlement from a Californian energy company in a scandal over contaminated water. Three decades later, she has a new target in her sights – and it’s global</p><p>When Erin Brockovich woke to find 30 emails from people from the same town, she realised something was going on. People email Brockovich all the time because of what happened in 1993, when she was instrumental in suing Pacific Gas and Electric Company (PG&E) on behalf of residents of the town of Hinkley

arXiv:2509.17932v2 Announce Type: replace Abstract: Large language models (LLMs) are prone to generating factually incorrect content, motivating methods for assessing truthfulness from internal model signals. While supervised probing approaches can be effective, they require labeled data and classifier training. Recent training-free methods avoid parameter optimization but rely on coarse activation statistics that provide limited insight into how truthfulness-related signals arise within the mod

arXiv:2606.27410v1 Announce Type: cross Abstract: The primary goal of Remote Sensing Image Change Captioning (RSICC) is to automatically generate descriptions of changes between remote sensing images captured at different time points. Existing models still rely on a single autoregressive generation paradigm, which tends to prioritize learning easily generated vocabulary over capturing discriminative differences between images. To address this, we reframe the training paradigm and propose a novel

arXiv:2606.27455v1 Announce Type: cross Abstract: We address the problem of inferring a directed network from nodal measurements generated by linear diffusion dynamics on the sought graph. Observations are modeled as the outputs of a graph convolutional filter, i.e., a polynomial (with unknown coefficients) of a local diffusion graph-shift operator encoding the latent graph topology, excited with an ensemble of independent graph signals with arbitrarily-correlated nodal components. Unlike prior

arXiv:2510.16492v4 Announce Type: replace Abstract: As Large Language Model (LLM) agents increasingly operate in complex environments with real-world consequences, their safety becomes critical. While uncertainty quantification is well-studied for single-turn tasks, multi-turn agentic scenarios with real-world tool access present unique challenges where uncertainties and ambiguities compound, leading to severe or catastrophic risks beyond traditional text generation failures. We propose using "q

arXiv:2511.03217v2 Announce Type: replace Abstract: Large language models (LLMs) excel in generating fluent utterances but can lack reliable grounding in verified information. At the same time, knowledge-graph-based fact-checkers deliver precise and interpretable evidence, yet suffer from limited coverage or latency. By integrating LLMs with knowledge graphs and real-time search agents, we introduce a hybrid fact-checking approach that leverages the individual strengths of each component. Our sy

arXiv:2602.10238v2 Announce Type: replace Abstract: The growing size of Large Language Models (LLMs) makes efficient inference challenging, primarily due to the memory demands of the autoregressive Key-Value (KV) cache. Existing eviction or compression methods reduce cost but rely on heuristics, such as recency or past attention scores, which serve only as indirect proxies for a token's future utility and introduce computational overhead. We reframe KV cache eviction as a reinforcement learning

arXiv:2606.27855v1 Announce Type: cross Abstract: Deep learning models for surface electromyography (sEMG) can benefit substantially from subject-specific (re-)calibration, since no sufficiently large and diverse datasets are available to train fully generic decoders. However, for user acceptance, the number of repetitions that can realistically be collected during calibration is severely limited, which increases the risk of overfitting and, in extreme cases, can even degrade performance compare

arXiv:2606.27863v1 Announce Type: cross Abstract: Demand forecasting at the bottom of a retail hierarchy requires predicting tens of thousands of correlated long-horizon series across products, stores, and regions. Modern systems must scale across massive catalogs, capture shared demand dynamics, and remain interpretable enough to be trusted. Classical statistical methods need a separate model per series and are hard to manage at scale; deep autoregressive models struggle as the joint state grow

arXiv:2606.27872v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models have demonstrated strong capabilities in robotic manipulation, but their performance degrades significantly in long-horizon tasks due to cumulative error propagation. This limitation largely arises from static feature fusion mechanisms that rely on fixed weights to combine visual, language, and action representations, preventing the model from adapting to different phases of task execution. To address this limi

arXiv:2606.27876v1 Announce Type: cross Abstract: Spatial intelligence is essential for low-altitude unmanned aerial vehicle (UAV) perception, collaboration, and navigation. However, existing UAV benchmarks often emphasize image-level recognition, single-view understanding, or narrow answer formats, leaving 3D spatial inference, multi-view collaboration, scene dynamics, and diverse task formulations insufficiently evaluated. To address these gaps, we introduce SpatialUAV, a real low-altitude UAV

arXiv:2606.27884v1 Announce Type: cross Abstract: Mixed-precision computation has been introduced in deep neural networks (DNNs) as an effective approach to reduce latency, energy consumption, and memory footprint. However, efficiently mapping mixed-precision networks onto multi-precision spatial architectures poses several challenges. These include determining the appropriate precision for each layer, balancing layer-wise accuracy sensitivity to quantization against architectural heterogeneity

arXiv:2606.27922v1 Announce Type: cross Abstract: Current multimodal reflection mechanisms for long video understanding predominantly rely on closed-loop self-reflection within internal parameters. Lacking objective external evidence, models are frequently trapped in blind confidence and often fail to correct errors. Furthermore, applying reinforcement learning to multi-stage reflection pipelines introduces severe policy coupling, which is exacerbated by a critical scarcity of dedicated training

arXiv:2606.27936v1 Announce Type: cross Abstract: The widespread collection of fine-grained location data by commercial data brokers creates a re-identification risk that is not widely recognised by the public. While prior research has established that mobility traces are highly unique and that individuals can, in principle, be identified from a handful of spatio-temporal points, such attacks have historically required significant manual effort from skilled analysts, limiting their practical sca

arXiv:2606.27939v1 Announce Type: cross Abstract: Protein language models are standard priors for biological sequence generation, but steering them toward explicit distributional design targets remains largely unexplored. We study a constrained protein generation problem in which sequences must match a desired amino-acid (AA) composition profile while preserving plausible sequence statistics and diversity. The motivating application is synthetic feed protein design, where the AA composition of d

arXiv:2606.27960v1 Announce Type: cross Abstract: Software engineering is an intellectually demanding, creative discipline that juggles a web of interdependent tasks to design, build, and assure the quality of increasingly complex systems. As our expectations from software soar - with demands spanning AI-driven products, pervasively distributed and cloud-native architectures, and deeply embedded cyber-physical environments - its complexity steadily increases. In response, a new wave of co-engine

arXiv:2606.27974v1 Announce Type: cross Abstract: Knowledge-based Visual Question Answering (KB-VQA) requires models to combine image understanding with external knowledge. Most prior methods use a fixed retrieve-then-generate pipeline with a pre-selected retriever and a static top-k setting, which is not adaptive during reasoning. We propose ProMSA, a progressive multimodal search agent for KB-VQA. Given an image-question pair, the agent iteratively chooses image search, text search, or stop, u

arXiv:2606.27978v1 Announce Type: cross Abstract: Pixel-space continuous-token autoregressive (AR) generation directly models images as sequences of raw pixel patches, avoiding discrete tokenization or a separately pretrained tokenizer. However, it faces coupled challenges: high-dimensional patch generation causes large single-step errors, and teacher-forced training creates a train--inference gap that makes these errors accumulate across AR steps. Existing fixes such as $x$-prediction and input

arXiv:2606.28027v1 Announce Type: cross Abstract: Neural video codecs have surpassed classical codecs in coding efficiency but remain impractical for deployment due to cross-platform incompatibility and high computational cost. Existing quantization-based solutions fail to produce deterministic results across diverse hardware platforms, leading to catastrophic decoding failures. We introduce MLVC, a hardware-robust neural video codec designed for practical cross-platform inference. The key idea

arXiv:2606.28039v1 Announce Type: cross Abstract: Demand for high-resolution satellite imagery has increased interest in super-resolution (SR) to bridge the spatial resolution gap between freely available missions such as Sentinel-2 and commercial systems like PlanetScope. Because no sensor provides true paired low- and high-resolution observations, SR models are usually trained on synthetically degraded data, creating a domain gap on real cross-sensor imagery. In this work, we provide the first

arXiv:2603.05121v2 Announce Type: replace Abstract: Speech Large Language Models route speech encoder representations into an LLM decoder that typically accounts for over 90% of total parameters. We study how much of this decoder capacity is actually needed for speech tasks. Across two LLM families and three scales (1-8B), we show that decoder redundancy is largely inherited from the pretrained LLM: text and speech inputs yield similar redundant blocks. We then measure excess capacity by pruning

arXiv:2606.28061v1 Announce Type: cross Abstract: Large language models (LLMs) have increasingly moved from standalone text generation systems to agents that invoke external tools, access environments, and execute multi-step tasks. However, conventional function-calling benchmarks mainly evaluate task completion and API correctness, while privacy evaluation benchmarks typically focus on final responses or privacy judgments. Neither perspective captures purpose-bound information flow across an ex

arXiv:2606.28065v1 Announce Type: cross Abstract: Understanding model predictions is essential for physical applications, where outputs often inform safety-critical decisions, such as structural load assessment, weather warnings, and clinical diagnosis. Shapley values satisfy many desirable properties as an attribution method, but their computational cost during inference hinders their practical use. Current amortized explainers, such as FastSHAP, are limited to homogeneous inputs, which is prob