Search: “reasoning”

arXiv:2603.17673v2 Announce Type: replace-cross Abstract: LLM agents are becoming increasingly important in the security domain, but leading systems are often closed-source, cloud-based, hard to repro

arXiv:2607.04751v1 Announce Type: cross Abstract: Big goals are hard to achieve all at once; breaking them into small steps is wiser. We present Trust Region Policy Distillation (TOP-D), which transfo

arXiv:2607.04484v1 Announce Type: new Abstract: Vision and vision-language models rely on high-level visual representations that are increasingly used across recognition, retrieval, and multimodal rea

arXiv:2607.05277v1 Announce Type: cross Abstract: Defenses that provide security guarantees against prompt injection attacks rely on strict isolation between trusted instructions and untrusted data. I

arXiv:2510.11098v5 Announce Type: replace-cross Abstract: Recent advances in large audio language models (LALMs) have greatly enhanced multimodal conversational systems. However, existing benchmarks r

arXiv:2511.20272v2 Announce Type: replace Abstract: While Multimodal Large Language Models (MLLMs) have become adept at recognizing objects, they often lack the intuitive, human-like understanding of

arXiv:2607.03835v1 Announce Type: cross Abstract: Converting a SPICE netlist into a human-readable schematic is a longstanding problem in electronic design automation: simulators and machine-learning

arXiv:2607.04962v1 Announce Type: new Abstract: Conspiracy theories commonly attribute important events to the actions of powerful and secretive actors. While computational research has largely focuse

arXiv:2607.03941v1 Announce Type: new Abstract: Recent advances in embodied AI have established robot foundation models (RFMs) as the dominant approach for generalist robotic systems to date. By lever

Hy3, the new 295B MoE model from @TencentHunyuan, is now free in Nous Portal for the next two weeks! It is focused on cost-effective agentic use, and particularly strong on coding, tool-calling reliab

Interesting stuff. And the visualization at the end is worth trying: https://www.neuronpedia.org/qwen3.6-27b/jlens New Anthropic research: A global workspace in language models. Of everything happenin

Most agentic retrieval demos assume clean, well-structured documents. Enterprise reality is often different, consisting of messy PDFs where critical information is buried across tables, figures, and c

One of the only times I remind people I have a PhD in computational neuroscience is when people without a neuroscience background say their model works 'like the brain.' In these cases, I put on my ne

// ReContext // Models now support 128K context windows and still fail to use evidence that is already in the prompt. Where is the gap? New paper introduces ReContext, a training-free inference harnes

'The frontier labs will keep owning discovery. Open source will increasingly own production.' Insightful take on a tired topic 'When you're running AI agents in production for customer service, latenc

As working with AI agents looks more like management, we may want to consider large-scale management training for the AI era. The US government actually did this once, & the WW2 Engineering, Science,

Fable prompting tips, based on all the guides Anthropic and its employees have given us. Three things you might want to do: 1) review your claude.md file against prompting best practices so you don't

spending the last week at @aidotengineer was awesome. too many great convos to cover them all, but jotted down some things that stood out: - Lots of discussion around open source models. I spoke with

The July 4th weekend All-In @theallinpod turned into a long argument about who owns the intelligence layer. The besties think enterprises just woke up to a trap they had been walking into, here's how

arXiv:2607.01287v1 Announce Type: cross Abstract: Accompanying a group of humans is an essential aspect of developing human-like social cognition in robots. However, human groups typically do not foll

arXiv:2607.02269v1 Announce Type: cross Abstract: Vision-Language Models (VLMs) have demonstrated immense promise in Spatio-Temporal Video Grounding (STVG). However, current evaluation protocols are l

arXiv:2510.04520v2 Announce Type: replace Abstract: Accurate auto-formalization of theorem statements is essential for advancing automated discovery and verification of research-level mathematics, yet

arXiv:2607.01387v1 Announce Type: cross Abstract: Recommender systems are vital in helping users navigate vast amounts of information, offering personalized suggestions and effective explanations for

arXiv:2607.02387v1 Announce Type: cross Abstract: NASA and its data centers hold thousands of geoscience datasets and tools like Worldview, Giovanni, the Science Discovery Engine, and Harmony. Finding

arXiv:2510.06288v4 Announce Type: replace Abstract: Today's AI models learn primarily through mimicry and refining, so it is not surprising that they struggle to solve problems beyond the limits set b

arXiv:2603.22435v2 Announce Type: replace-cross Abstract: 'Code-as-Policy' considers how executable code can complement data-intensive Vision-Language-Action (VLA) methods, yet their effectiveness as

arXiv:2607.01936v1 Announce Type: cross Abstract: Learning causal models from high-dimensional data is a significant challenge, particularly in real-world settings where violations of core assumptions

Don't train the model, evolve the harness. I read a brilliant blog post from Hugging Face where they took a frozen open model scoring 0% on a hard legal agent benchmark, left its weights alone, and le

arXiv:2603.18315v2 Announce Type: replace-cross Abstract: Traditional reinforcement learning (RL) methods rely on manually engineered rewards or sparse collision signals, which fail to capture the ric

arXiv:2607.02007v1 Announce Type: new Abstract: Large language models now score near ceiling on general benchmarks, but these aggregate measures reveal little about how models behave within single dis

arXiv:2607.02118v1 Announce Type: new Abstract: Scientific Fitness Coaching (SFC) is typically delivered by human professionals, making it costly and inaccessible to many. While recent advances in Lar

Excited to share our paper, “Learning Multi-Agent Coordination via Sheaf-ADMM” to be presented at #ICML2026 Blog: https://pub.sakana.ai/sheaf-admm/ Most AI models process information as one giant, mon

arXiv:2603.13191v2 Announce Type: replace-cross Abstract: While large language models (LLMs) have transformed AI agents into proficient executors of computational materials science, performing a hundr

arXiv:2607.02329v1 Announce Type: new Abstract: Autonomous-research agents have demonstrated end-to-end LLM automation in machine-learning sandboxes where execution provides calibration. Frontier phys

arXiv:2606.22737v2 Announce Type: replace Abstract: Before letting an agent operate over real context, can you prove it used the right evidence? GroundEval turns that question into a deterministic tes

arXiv:2606.14249v2 Announce Type: replace Abstract: AI agent performance depends critically on the runtime harness, comprising the prompts, tools, memory, and control flow that mediate how a model obs

arXiv:2607.02467v1 Announce Type: cross Abstract: Whether pairing people with AI helps or hurts is usually reported as a single average effect. Using a real-money prediction market (Polymarket) as an

arXiv:2607.02383v1 Announce Type: new Abstract: LLM-based retrieval-augmented generation (RAG) is increasingly used for automated fact-checking (AFC) and related tasks. By grounding LLM outputs in ret

arXiv:2504.02327v2 Announce Type: replace Abstract: Natural Language to SQL (NL2SQL) aims to translate natural language queries into executable SQL statements, offering non-expert users intuitive acce

arXiv:2607.01740v1 Announce Type: new Abstract: Public LLM leaderboards optimise for global average performance and do not capture the specific cognitive demands of financial-services work: a model th

arXiv:2607.01813v1 Announce Type: cross Abstract: Evaluation benchmarks are essential for assessing vision-language models (VLMs), but most multimodal benchmarks are static, making them vulnerable to

arXiv:2607.01814v1 Announce Type: new Abstract: Traditional Chinese Medicine (TCM) diagnosis, particularly through tongue inspection, faces persistent challenges in subjectivity and reproducibility. T

arXiv:2607.01982v1 Announce Type: cross Abstract: Using molecular large language models (LLMs) as a unified framework for understanding molecular structures and functions is emerging as a new trend in

arXiv:2607.02423v1 Announce Type: cross Abstract: Active Few-Shot Learning (AFSL) adapts LLMs to specialized domains by identifying the most valuable unlabeled samples for annotation and use as few-sh

arXiv:2607.02510v1 Announce Type: new Abstract: Despite alignment training, LLMs remain prone to generating unsafe outputs at deployment time. Monitoring outputs online and raising an alarm when safet

OpenAI reportedly has floated giving the U.S. government a 5% stake in the company, perhaps the start of a series of such stakes in other AI companies as well. This no doubt has traditional anti-indus