Skip to content
AI NEWS JOURNAL
AI, IN PERSPECTIVE

Search AI news

Search headlines, summaries and articles to connect the dots.

Newest first

ReKV: Overlapping Quantized KV Transfer and Suffix Recomputation

arXiv:2611.08770v1 Announce Type: new Abstract: Prefill--decode disaggregation moves a prompt's key--value (KV) cache between workers before the first token can be generated. Reducing this handoff to either cache compression or recomputation misses a useful choice: which layers to transmit, at what precision, and which layers to execute again on the decode worker. ReKV transfers a quantized prefix KV cache and one BF16 boundary state, then reconstructs a contiguous suffix while the prefix is in flight. In a model-shaped synthetic benchmark on cross-node H100s, the Q8 configuration meeting separately evaluated quality criteria reduces 8K handoff latency by $2.36\times$, $2.84\times$, and $3.65\times$ over BF16 transfer for Qwen2.5-7B, Qwen2.5-3B, and Qwen3-8B. It is also faster than the best measured transfer-only or recompute-only action at those points. A separate real-weight BF16 recomputation experiment reduces prefill-to-token-return time by $1.16$--$1.63\times$. The preferred action changes with prompt length and link bandwidth.

Show HN: Control Layer Demo for AI Coding Agents

Letting coding agents (such as Claude Code) work autonomously on repos on a daily-basis, brings a high level of uncertainty. For those who won't let the agent decide alone whether it can create/modify or delete any folder on their repos; I wanted to be able to trace actions performed and being able to let the agent continue his work autonomously, or ask him to request a human approval according to some criteria defined. After investigations, the cleanest way to do it was according to me to develop a deterministic control layer between the LLM and the work of the agent that would track and catch potential non-respect of pre-defined rules. How? by first setting up approval rules (Pass/requires approval/refused) before the work of the agent, and afterwards controlling cryptographic SHA 256 verifications, etc. This allows to compare the before-after work of the agent. I have put in my public GitHub repo a demo that you can launch easily and observe different scenarios behaviors like applied authorized modification, protected file waiting for Human approval, or outside perimeter authorization modification refused. The demo is offline, does not involve the official runtime, in native node.js and without dependencies. It should still allow you to understand the core principles. I would like to know: how do you implement and handle safety boundaries? Especially if you let agents work on critical data/sensitive codebases? With which guardrails? Comments URL: https://news.ycombinator.com/item?id=50422833 Points: 1 # Comments: 0

Transporting the Past: An Optimal Transport View of Backtracking Counterfactuals

arXiv:2611.08682v1 Announce Type: new Abstract: We study backtracking counterfactual inference for Structural Causal Models (SCMs) through the lens of optimal transport. In SCMs, a factual observation generally induces a posterior distribution over exogenous variables, making counterfactual inference inherently distributional. We introduce two complementary frameworks that transport this posterior onto the set of exogenous configurations satisfying a desired counterfactual query. The first constructs a natural pointwise projection and is suitable when the transported exogenous distribution can be used directly. The second enforces mutual independence among counterfactual exogenous variables, so that the transported variables remain interpretable as genuine exogenous noises of the same SCM. We show that both frameworks are consistent with the original backtracking counterfactual principle, and we propose marginal, penalty-based, and hard-constrained solvers for finite-posterior settings. Synthetic experiments on non-bijective SCMs illustrate the geometry of the problem and reveal trade-offs among transport cost, counterfactual constraint satisfaction, and preservation of exogenous independence.

How Greenlight Financial built an autonomous SOC solution on Amazon Bedrock AgentCore

Greenlight Financial resolves more than 90% of up to 4,000 monthly security and risk cases without human intervention using Luna, a tiered multi-agent system on Amazon Bedrock AgentCore. This post covers the architecture, episodic memory for continuous learning, and the expansion from SOC operations into financial risk triage.

Serverless MCP server for AI agent data lake access on Amazon ECS Fargate

Learn how to build a serverless Model Context Protocol (MCP) server on Amazon ECS with AWS Fargate that gives AI agents secure, metadata-driven access to an enterprise data lake. Exposing a new dataset requires only a JSON file, with no per-dataset integration code.

Show HN: Combining Bursts of Images

An approach of taking many photos to enhance one final image is a technique called "stacking". Stacking alone doesn't account for everything so this is a take that helps get better alignment of images, allows comparisons, and drop bad images that dont match. Its not super resolution but it feels like when you look at it. it reduces the noise and makes it look more organic shot with less expensive hardware. enjoy! Comments URL: https://news.ycombinator.com/item?id=50401218 Points: 1 # Comments: 0

Gemini Omni 1.2 Flash: Fast and cost-effective with longer clips

Google DeepMind has launched Gemini Omni 1.2 Flash, introducing native 15-second video generation, improved reference adherence, stronger consistency across frames, multi-turn conversational editing, more natural speech, and cinematic camera movements. The model is now generally available via the Gemini API and Gemini for business, priced at $0.07 per second for 720p video—30% lower than its predecessor. It supports scalable workflows, including draft generation at lower resolutions, iterative edits, and upscaling to 1080p or 4K. Use cases include product explainers, social ads, virtual try-ons, and branded content, with applications across Google AI Studio, Google Cloud, Google Vids, Flow, the Gemini app, and Google Ads.

New NVIDIA Cosmos 3 Open Models for Interactive World Simulation

NVIDIA has introduced four new open-weight checkpoints for its Cosmos 3 model family—Cosmos3-Nano-Transfer-Auto, Cosmos3-Nano-Sim-Auto, Cosmos3-Nano-Sim-Transfer, and Cosmos3-Nano-Sim-Bimanual—enabling interactive world simulation and multi-sensor, multi-view driving scene generation. Each model is a post-trained variant of Cosmos 3 Nano with 16B trainable parameters and is based on OpenMDW1.1, allowing commercial and non-commercial use. The checkpoints support applications in robotics and autonomous driving by generating synchronized camera and LiDAR data, enabling streaming simulations, translating simulator controls into photorealistic video, and predicting visual outcomes from robot actions. They are designed to be adapted using custom data and training recipes, with support for offline data generation, causal distillation, and interactive rollouts. Resources include technical reports, demos, and notebooks for inference and adaptation.

Show HN: Umbel – Remote-control agent CLIs over tmux

Been using this in all my sessions the last months. Opus 5.5 (over MCP) spawns Deepseek flash agents - and codex for audits etc. Very curious to hear what the HN crowd thinks/makes of it. Comments URL: https://news.ycombinator.com/item?id=50383189 Points: 1 # Comments: 0

Synthesizing symbolic models of human and machine reasoning at scale

arXiv:2611.07088v1 Announce Type: new Abstract: Symbolic models offer a powerful account of cognition, but their scope has historically been limited by the need for hand-engineering and computational tractability. We show how advances in artificial intelligence (AI) can scale symbolic modeling and help overcome these engineering limitations. We introduce Behavior-Conditioned Symbolic Synthesis (BCSS)---an approach for automated search of interpretable, deterministic symbolic programs that verifiably reproduce behavior---and apply it to model both human and machine behavior across hundreds of challenging reasoning problems. Analyzing these programs reveals that both people and machines tend to oversimplify, adopt rules that are partial explanations for their observations, and misinfer dependencies between problem features. These shared signatures are modulated by available resources, with reasoning effort predicting the extent to which machines simplify and explain their observations. Together, these results point to underlying mechanisms reminiscent of satisficing in both people and machines, and illustrate how BCSS offers a general approach for interpreting and comparing rule-based reasoning of humans and black-box AI systems.

WeLM-Image: A Unified Image Generation Model towards Visual Intelligence

arXiv:2611.07392v1 Announce Type: new Abstract: We present \textbf{WeLM-Image}, a unified image foundation model that integrates multimodal understanding into text-to-image generation, instruction-guided editing, and multi-reference synthesis. Designed for single-GPU deployment in WeChat's large-scale consumer applications, the model processes text, reference images, and target latents as one interleaved sequence. A Mixture-of-Transformers (MoT) backbone combines shared attention with independent understanding and generation parameters to limit optimization interference. Equipping each path with Mixture-of-Experts (MoE) modules provides sparse capacity, with 31.7B stored parameters and approximately 5.4B generation parameters activated per target latent token. A generation-oriented data engine combines a knowledge-grounded visual vocabulary with large-scale visual clustering to balance semantic coverage and editing tasks, while training-time evaluation guides iterative data refinement. Bilingual annotations and verified editing supervision support progressive training up to native 1.5K resolution. We optimize generation and editing separately with rubric- and checklist-based rewards, then consolidate the specialized policies through noise-aware on-policy distillation. In the reported overall comparisons, the 1536-resolution model ranks first among the evaluated open-source models on PRISM-Bench, Qwen-Image-Bench, and UnicBench. MeanFlow-based few-step distillation yields WeLM-Image-Turbo, which generates a $1024\times1024$ image in 4.98\,s on a single NVIDIA H20 GPU using eight guidance-free steps.

A Closer Look at Phonon TTS. Launching High-Quality On-Device Speech That Tops the Benchmarks.

Phonon, a small text-to-speech model that runs entirely on the user's device without requiring internet access, has been released with support for English, French, German, Spanish, and Portuguese. Built on the Pocket TTS lineage using flow matching, Phonon achieves high-quality audio in compact sizes (65 MB and 75 MB), with the 31m version licensed under Apache 2.0 for commercial use. It excels in intelligibility and speed, ranking first in German and Spanish on Hugging Face's Open TTS Leaderboard and leading in CPU streaming benchmarks. The model supports voice cloning, custom voices via Gradium Studio, and integrates with Hugging Face Transformers. It is available in multiple formats and backends, including WebAssembly, GPU, and NPU, enabling deployment across phones, laptops, single-board computers, and browsers. Over 200,000 downloads were recorded in its first two weeks, with active developer adoption through tools like ExecuTorch, Micdrop, and Spaces. Future developments include support for additional languages, smaller models, and custom vocabulary fine-tuning.

Cognitive Bias Signatures in Search-Equipped LLM Agents: A Stage-Resolved Analysis

arXiv:2611.05720v1 Announce Type: new Abstract: Search can help LLM agents correct mistakes, but their starting conditions may shape which evidence they find and use. Drawing on cognitive-bias paradigms, we develop a controlled search harness to examine result order, records of prior work, and initial hypotheses while holding each question and document collection fixed. Matched controls distinguish order effects from differences in document composition and the effects of failed work from context length. Across 10 models and 1,000 synthetic search worlds, we track queries, retrieved documents, document openings, and final answers. Our comparisons suggest that final answers are more sensitive to initial evidence order during active search than when all evidence is supplied in full. Dead-end histories can reduce strategy switching relative to matched neutral histories, indicating an effect beyond context length. Under wrong hypotheses, stage measurements distinguish reduced retrieval or opening from errors that persist after counter-evidence is opened. Exploratory interventions show that opening counter-evidence more often can coincide with lower accuracy, while accuracy can improve with little change in opening.

Dharma Field Extractor: Any Document In, Structured Fields in JSON Out

The Dharma Field Extractor, a 2-billion-parameter model, specializes in extracting specific fields from diverse document types—such as invoices, IDs, and medical forms—into structured JSON format. It outperforms general-purpose frontier LLMs in accuracy and reduces cost by up to 147 times when self-hosted, demonstrating that domain-specialized small models can surpass large general-purpose models on well-defined tasks. The model supports both open extraction (detecting fields without a schema) and schema-directed extraction (returning exact field names and types), with training data spanning English, Brazilian Portuguese, and multilingual documents. Benchmarks show superior F1 scores and valid JSON output across real-world datasets like SROIE and FUNSD, reinforcing the thesis that specialization beats scale for enterprise document processing.

SRRM: A Retrieval-Reconstruction Benchmark for Evaluating Long-Context Memory in Small Language Models

arXiv:2611.05386v1 Announce Type: new Abstract: Small Language Models (SLMs) are increasingly deployed on consumer-grade hardware, yet their ability to preserve information across progressively longer contexts remains insufficiently understood. Existing benchmarks primarily evaluate retrieval-oriented tasks, providing limited insight into whether successful retrieval reflects comprehensive long-context memory preservation. To address this limitation, this paper introduces Simple Retrieval-Reconstruction Memory (SRRM), a lightweight and fully reproducible benchmark that jointly evaluates Targeted Retrieval and Global Information Reconstruction. This study also proposes the Memory Preservation Ratio (MPR), a benchmark-level metric that quantifies the relationship between retrieval and reconstruction performance. SRRM was evaluated using a synthetic dataset on three Qwen2.5 models (0.5B, 1.5B, and 3B parameters) across context loads ranging from 10 to 100 factual statements. Experimental results show that reconstruction performance deteriorated substantially faster than retrieval accuracy as contextual information increased, indicating that retrieval alone is insufficient to characterize long-context memory preservation. Among the evaluated models, Qwen2.5-1.5B achieved the highest retrieval accuracy (86.7%), reconstruction performance (45.7%), and MPR (0.53), outperforming the larger Qwen2.5-3B model (72.2%, 18.5%, and 0.26, respectively). These findings demonstrate that increasing parameter count does not necessarily improve long-context memory preservation. Overall, SRRM provides a reproducible and extensible benchmark for systematically evaluating long-context memory while revealing complementary memory behaviors that remain hidden under retrieval-only evaluation frameworks.