Skip to content
AI NEWS JOURNAL
AI, IN PERSPECTIVE

Search AI news

Search headlines, summaries and articles to connect the dots.

Newest first

KV-LTM: Persistent Per-Session KV Caches as Long-Term Memory for Personal Assistants

arXiv:2611.21998v1 Announce Type: new Abstract: Personal assistants accumulate months of conversations, but today's long-term memory stacks put derivation or selection between a question and its evidence. In a system-level comparison on LongMemEval, two commercial assistants scored well below the same model reading the history offline [1], consistent with, but not isolating, content loss between write and read. We propose KV-LTM: a memory tier storing each session's raw text plus its KV cache under a fixed reader model, answering by restoring cached sessions and appending the question. KV-LTM is complementary to, not a replacement for, retrieval and indexing. Text is the memory of record; the cache is a recompute-free prefill accelerator, degrading gracefully to text. We specify the design: storage record, budget-derived retention, append-only read path, and bounded aggregation. We size it analytically: across pinned BF16 full-attention (GQA) and latent-cached MLA configurations, per-token state spans approximately 31K-328K bytes (the pinned KDA+MLA hybrid differs: approximately 8.1K bytes/token plus a fixed per-session recurrent term), giving approximately 0.34-3.6 TB per user-year at a declared 30K-token/day scenario, against approximately 49 MB of raw text; per-query cost is modelled as regime-conditional transfer, attention, decode, aggregation, and compute. We define an evaluation protocol needing no persistent-cache implementation for logical QA evaluation: per-session prefill emulates valid cache reuse under the same model/input-state contract and ideal arithmetic, with named arms, controlled baselines, and a preregistered reader-model gate; a separate minimal real-cache replay tests numerical equivalence and measures the cost axis, with outcomes not assumed. KV-LTM trades storage for avoided re-prefill compute and write-time fidelity, within a declared retention budget.

On Formal Counterfactual Explanations

arXiv:2611.22010v1 Announce Type: new Abstract: Counterfactual explanations are often seen as counterfactual conditionals under closest-world semantics, requiring one to reason over all minimally distant counterexamples. In contrast, formal XAI typically relies on feature-based abductive explanations (AXps) and contrastive explanations (CXps), which admit efficient computation and a minimal hitting-set duality. This paper bridges these perspectives by (i) characterizing closest-world counterfactual explanations explicitly as a disjunctive normal form over all minimally distant counterexample instances, and (ii) introducing conjunctive counterfactual explanations (CFXps), a subclass of counterfactual antecedents restricted to conjunctions of feature literals that guarantee prediction change. The paper establishes structural results connecting CXps, AXps and CFXps: every CXp induces an AXp for its witness instance, and every CXp together with a witness AXp induces a CFXp whose support is precisely determined by that AXp. These theoretical results yield a modular approach for computing CFXps using existing AXp/CXp machinery. Experiments on a representative set of benchmarks using tree ensembles and neural networks show that CFXp extraction is practically efficient. Compared with popular point-based counterfactual explainers, CFXps generally require no more feature changes and can guarantee the counterfactual prediction beyond a single instance.

Parallel web search is free on AI Gateway and eve through January 31

Parallel web search in fast mode is free on AI Gateway until January 31, 2027, and is now the default for eve agents. It works with any AI Gateway model without a separate API key, returning links and excerpts with filtering options by website or date. The feature is automatically used in eve's web_search tool when models are served via AI Gateway. Developers can enable it via the AI SDK using gateway.tools.parallelSearch({ mode: 'fast' }). Fast mode includes Zero Data Retention by default; other modes are billed at standard rates.

Reason over time-series data with OpenTSLM on Amazon SageMaker JumpStart

OpenTSLM, an open source family of time-series language models, is now available in Amazon SageMaker JumpStart. See how Nexthink and Aionic Labs apply it to proactive IT support, then deploy OpenTSLM to a real-time endpoint and fine-tune it on your own data with TimeNet and a SageMaker training job.

Responsible AI for agentic AI: A phased roadmap for platform and application teams

Enterprises adopting agentic AI at scale need platforms that embed responsible AI from the start. This post gives platform and application teams a three-phase roadmap (governance foundation, pilot deployment, and evaluation and hardening) that maps the AWS Responsible AI Lens pillars to each phase, with guidance on when to revisit each one.

Show HN: An ASCII cell aquarium for your terminal

I made a virtual microscope slide for your terminal (inspired by asciiquarium and cbonsai). Cells divide, immune cells chase bacteria, and bacteriophages occasionally cause trouble. You can try it with:$ pipx install terminal-cells$ cellsI'd appreciate any suggestions for additional animations to add, especially entities or events that would be visually distinct given the graphical limitations. Any cell biologists in the house? Comments URL: https://news.ycombinator.com/item?id=50541047 Points: 1 # Comments: 3

Show HN: JIT Execution of LLVM IR Modules

We made it a little easier to execute modules in the LLVM dialect of MLIR, thought this might be interesting for the HN community thinking of JITting Python, as an alternative to writing the LLVM IR in llvmlite directly. Comments URL: https://news.ycombinator.com/item?id=50534847 Points: 2 # Comments: 1

The Silent Consensus

Dean Ball, Head of Strategic Futures at OpenAI, examines how more capable AI could reshape society and how people can retain control of their institutions.

Track Copilot code review outcomes in Repository insights

You can now see how Copilot code review feedback lands in each repository. A new “Repository insights” page, the first item in a repository’s Insights tab, brings Copilot code review… The post Track Copilot code review outcomes in Repository insights appeared first on The GitHub Blog.

v1.20.0

Dify 1.20.0 introduces support for exporting traces to any OpenTelemetry collector, enables malware scanning of uploaded knowledge files, adds real-time Agent tool activity visibility in Chatflows, and allows automation via personal access tokens. New Human-in-the-Loop (HITL) features integrate approvals into messaging tools like Slack and Teams, while configuration changes include updated defaults for security and integration settings. The release also includes database migrations, improved performance, and bug fixes related to file handling, authentication, and worker stability.

Vercel Blob now supports TTLs for automatic deletion

Vercel Blob now allows automatic deletion of blobs using a time-to-live (TTL) feature, where blobs are deleted automatically between 1 to 730 days after upload. TTL can be set during write operations like put(), copy(), putImage(), multipart uploads, and presigned URLs. The expiresAt timestamp is included in all write results and head() responses. TTL can be modified or removed later using setTtl() without re-uploading. CLI commands support --ttl-days and --clear-ttl flags. Setting a TTL incurs one Advanced Operation per 60 days, with costs scaling up to 13 for the maximum 730-day TTL. Deletion and clearing TTL are free. Usage can be tracked via the Observability tab.

Your complete guide to AWS Quick at AWS re:Invent 2026

AWS Quick gives your whole organization AI that knows your business, respects your security rules, and acts on your behalf. At AWS re:Invent 2026, customers including Allianz, Cox Automotive, and Southwest Airlines share how they govern, extend, and scale Quick. Use this guide to find the sessions that fit your goals.

b12475

The Llama Chat UI has been upgraded to SvelteKit 3, integrating its built-in service worker for improved update detection and offline functionality. The migration includes updates to Vite 8, TypeScript 6, Storybook 10, and ESLint 10, along with the replacement of deprecated packages and resolution of npm audit issues through dependency overrides, including pinning minimatch to version 11.0.5 to address a security vulnerability. Changes also align with current Svelte patterns, such as using $state instead of stores and adopting new syntax for actions and class bindings.

Decision-1.1 now available on AI Gateway

Decision-1.1, an improved version of Microsoft's decision model, is now available on AI Gateway with enhanced accuracy, better calibration, and faster performance. It provides structured outputs—such as Boolean probabilities, choice selections with confidence scores, and numerical ratings—for tasks like routing support requests, selecting models, workflow steps, or grading AI responses. Multiple questions can be processed in a single request, and the model maintains consistent probability tracking to enable threshold-based actions. It uses the same request format as Decision-1 but requires updating the model ID to 'microsoft/decision-1.1'. Accessible via AI SDK, OpenAI-compatible Decisions API, or AI CLI, it supports logging, budget tracking, and fallback mechanisms. Documentation and examples are available for integration.