Search: “context”

arXiv:2610.01511v2 Announce Type: replace Abstract: Most preference optimization methods, such as Direct Preference Optimization (DPO), apply preference supervision at the response level, although aut

After some recommendations for hardware (I'm starting from nothing), in the region of £6-7k. I will be mainly looking to use it for coding and have found Deepseek V4.1 Flash or Qwen3.8 27b good and so

Introducing North 2. Enterprises have been compromising on agentic AI for years: platforms that aren't scalable, models that aren't sovereign, control without enterprise-grade security. That ends toda

Finally finished my fork: https://github.com/IceFog72/ik_llama.cpp Nothing else I wanted to add/try currently works In short, it now has: Bandwidth-adaptive CPU/GPU execution Shared expert-residency i

arXiv:2610.03564v1 Announce Type: new Abstract: Financial AI agents must do more than retrieve facts: investment workflows require correct quantitative execution, reliable use of procedural resources,

arXiv:2610.02774v1 Announce Type: new Abstract: Long-form generators for music, motion and video produce sequences chunk by chunk, with each chunk generated by iterative denoising while rewards are de

arXiv:2610.03065v1 Announce Type: new Abstract: Learning control from action-free recordings is challenging because intervention effects are unobserved and policies may exploit errors in reconstructed

arXiv:2610.03636v1 Announce Type: cross Abstract: Camera-controlled video models are rapidly advancing toward long generation horizons and complex camera control. A key failure mode is 3D inconsistenc

arXiv:2610.02627v1 Announce Type: new Abstract: An email assistant should not complete less work simply because a user phrases the same request differently. Yet most benchmarks test each task with onl

arXiv:2610.03093v1 Announce Type: new Abstract: Standard autoregressive (AR) models process high-level task instructions, state history, and transient tokens within a single shared sequence of tokens.

arXiv:2610.02345v1 Announce Type: new Abstract: A family of tabular anomaly detectors trains a neural map toward a fixed target under squared-error loss and scores anomalies by the test-time residual;

arXiv:2605.12276v2 Announce Type: replace Abstract: Vector geospatial data represent the world as discrete geoentities, such as roads, buildings, and points of interest, each with semantic attributes,

Hey all. We've spent the last weeks getting Qwen3.8-Flash-Next (125B MoE, 6B active) to run properly on one AMD Strix Halo box (Ryzen AI Max+ 395, 128 GB). Tonight we're releasing both the 95 GB EXL3

arXiv:2605.06639v2 Announce Type: replace-cross Abstract: We introduce Recursive Agent Optimization (RAO), a reinforcement learning approach for training recursive agents: agents that can spawn and de

arXiv:2610.03073v1 Announce Type: cross Abstract: Security workflows need models that turn complex observations and explicit policies into decisions. System One models introduced by Jev return typed p

arXiv:2610.03268v1 Announce Type: new Abstract: Touche 2026 extends causality extraction to counter-causal claims: news sentences whose surface form appears causal but whose meaning denies the causati

arXiv:2610.03029v1 Announce Type: new Abstract: Gene set analysis is a cornerstone of functional genomics, yet it remains labor-intensive and heavily dependent on manual curation and expert biological

tl;dr: I created a fully-local open-source full-duplex voice agent that rivals GPT-Live on some benchmarks. It uses Voxtral Realtime with a turn-taking head, a microturn-finetuned Gemma 4 12B and Bree

arXiv:2610.03215v1 Announce Type: new Abstract: StanceEval 2026 is the second edition of the StanceEval shared task series on stance detection in Arabic social media text. Stance detection aims to ide

arXiv:2610.02248v1 Announce Type: new Abstract: Operational land surface forecasting systems built on Mamba-family Structured State Space Models absorb non-stationary confounding events (unrecorded ir

arXiv:2610.02722v1 Announce Type: new Abstract: Structural connectivity (SC) and functional connectivity (FC) provide complementary information on interactions between brain regions and are widely use

arXiv:2610.02516v1 Announce Type: cross Abstract: Zero-shot classifiers are useful for routing user requests to specialized LLM tasks, but scoring every request against a large candidate set is expens

arXiv:2610.03124v1 Announce Type: cross Abstract: Open-weight language models can be downloaded, modified, and deployed beyond their developers' control, limiting the effectiveness of centrally enforc

arXiv:2610.02247v1 Announce Type: cross Abstract: Artificial intelligence increasingly serves as a natural-language interface to complex technical systems, letting people accomplish sophisticated task

arXiv:2610.02736v1 Announce Type: cross Abstract: A compressor can keep the facts of a dialogue and still drop the turn that changed them. A user corrects a price, reverses a choice, or adds a constra

arXiv:2610.02478v1 Announce Type: new Abstract: Reinforcement learning for large language models typically maximizes expected return, adding up the probabilities of all successful trajectories. Howeve

arXiv:2610.03025v1 Announce Type: new Abstract: What makes a short story gripping; a news article newsworthy; or a math proof elegant? These constructs resist articulation or verification; their meani

arXiv:2610.02713v1 Announce Type: new Abstract: Most KV-cache compression methods classify attention heads once, either offline or during prefill, and keep this classification fixed throughout generat

arXiv:2610.02508v1 Announce Type: new Abstract: World action models (WAMs) have emerged as a promising paradigm for robotic control by jointly predicting future visual dynamics and actions from an ini

​ From the first LLaMA 33B I knew I wanted that magic-like intelligence locally, mine, so nobody could take it away when I needed it. I bought a 3090 for my home PC. Then LLaMA 65B appeared and I was

HuggingFace just closed a major gap in harness engineering! (the ultimate guide to multi-harness RL) the same open-weight model can perform well inside one harness, then lose accuracy or produce inval

A while ago I asked here how to turn ~5 million court decisions into structured graphs without running an expensive LLM on every document thanks for the advice . I went with the 'small extractor + cla

First of all, thank you for reading. I trained a small MoE model completely from scratch (no external base weights) and wanted to share the results + a couple of lessons. Apex-2 - Architecture: Decode

I tried both these models on a few small agentic tasks with tools (i.e. 'What's the weather like today?', etc.). They're both pretty solid at using web search to find answers despite being small model

Auto-generated synthesis of 29846 entries about arxiv-cs-ai

Auto-generated synthesis of 3157 entries about techmeme

Banger paper from Meta Superintelligence Labs. They find something super interesting and unexpected. (bookmark it) Base models with a light harness often solve more agentic tasks than their RL post-tr

So this is maybe a niche problem, but at my job I work on a huge Python codebase and every time I change some shared function I'm basically playing roulette. grep tells who mentions it in the code bas

Yann LeCun's (@ylecun ) latest talk at ETH Zürich Scaling LLMs to reach AGI is 'impossible' A large language model is trained on about 30 trillion tokens, which is roughly 10^14 bytes of text and woul

arXiv:2610.00369v1 Announce Type: cross Abstract: Laurito et al. (PNAS 2025) showed that large language models choosing between two descriptions of the same product, paper or film prefer the descripti

arXiv:2610.01861v1 Announce Type: new Abstract: Natural language descriptions can provide rich semantic representations of audio-visual urban scenes, yet datasets that jointly describe both auditory a

arXiv:2610.00663v1 Announce Type: new Abstract: Backdoored large language models (LLMs) can behave normally on benign inputs while producing attacker-specified outputs under hidden triggers. Existing

arXiv:2610.00197v1 Announce Type: new Abstract: We investigate automated rewards for training language models in conversational humor, focusing on reward exploits and countermeasures. Two approaches a

arXiv:2610.00499v1 Announce Type: new Abstract: As diffusion large language models (dLLMs) become more capable, they are moving from research settings to real-world extit{serving}, where request manag

arXiv:2610.01207v1 Announce Type: new Abstract: When training large language models with reinforcement learning, terminal rewards provide little guidance about which steps matter. Common methods for a

arXiv:2603.18480v2 Announce Type: replace-cross Abstract: Inferring human engagement from gameplay video is important for game design and player-experience research, yet it remains unclear whether vis