Search: “models”

SPICE: Submodular Penalized Information-Conflict Selection for Efficient Large Language Model Training

arXiv:2601.23155v2 Announce Type: replace-cross Abstract: Information-based data selection for instruction tuning is compelling: maximizing the log-determinant of the Fisher information yields a monot

researcharxiv-cs-ai

Human

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Safety

Synthetic Data for any Differentiable Target

DGX agent

arXiv:2604.08423v1 Announce Type: new Abstract: What are the limits of controlling language models via synthetic training data? We develop a reinforcement learning (RL) primitive, the Dataset Policy G

safetyarxiv-cs-cl

10 Apr 2026

Through the Magnifying Glass: Adaptive Perception Magnification for Hallucination-Free VLM Decoding

arXiv:2503.10183v4 Announce Type: replace Abstract: Existing vision-language models (VLMs) often suffer from visual hallucination, where the generated responses contain inaccuracies that are not groun

researcharxiv-cs-cv

TraceSafe: A Systematic Assessment of LLM Guardrails on Multi-Step Tool-Calling Trajectories

arXiv:2604.07223v1 Announce Type: cross Abstract: As large language models (LLMs) evolve from static chatbots into autonomous agents, the primary vulnerability surface shifts from final outputs to int

Weakly Supervised Distillation of Hallucination Signals into Transformer Representations

arXiv:2604.06277v1 Announce Type: new Abstract: Existing hallucination detection methods for large language models (LLMs) rely on external verification at inference time, requiring gold answers, retri

Weaves, Wires, and Morphisms: Formalizing and Implementing the Algebra of Deep Learning

arXiv:2604.07242v1 Announce Type: new Abstract: Despite deep learning models running well-defined mathematical functions, we lack a formal mathematical framework for describing model architectures. Ad

researcharxiv-cs-lg

What Drives Representation Steering? A Mechanistic Case Study on Steering Refusal

arXiv:2604.08524v1 Announce Type: cross Abstract: Applying steering vectors to large language models (LLMs) is an efficient and effective model alignment technique, but we lack an interpretable explan

safetyarxiv-cs-cl

AI on the couch: Anthropic gives Claude 20 hours of psychiatry

model-releasesars-technica

As part of the evaluation of its Claude Mythos model, Anthropic engaged a clinical psychiatrist for approximately 20 hours of evaluation sessions, describing Mythos as 'the most psychologically se...

9 Apr 2026

With closed agent platforms like Claude Managed Agents, your agent's memory belongs to them, not you. It's locked behind their API. Agent in…

model-releasesharrison-chase--x

With closed agent platforms like Claude Managed Agents, your agent's memory belongs to them, not you. It's locked behind their API. Agent infra should be open: open harness, open memory, model agnosti

9 Apr 2026

Today, our group at @Mila_Quebec and the lab of @francesarnold at @Caltech just released a new paper I contributed to, exploring how multimo…

model-releasesyoshua-bengio--x

Today, our group at @Mila_Quebec and the lab of @francesarnold at @Caltech just released a new paper I contributed to, exploring how multimodal generative modeling could accelerate protein sciences! ⬇

8 Apr 2026

GLM 5.1 is now LIVE in Atomic Chat SOTA for code & chat – now runs locally with TurboQuant Thanks to @zai_org for open-sourcing this frontie…

model-releaseszhipu-ai--x

GLM-5.1 is Z.ai's (zai-org) next-generation open-source flagship model for agentic engineering, achieving state-of-the-art performance on SWE-Bench Pro and significantly outperforming its predecess...

7 Apr 2026

DeepSeek v4.1 Flash 1017 for Dwarfstar (ds4)

model-releasesr-localllama

Hey y'all, just following up on my PR to get the new vision tower for 1017 working in ds4 with the quant I promised (the language model loads already, but vision needed my PR). I built to the exact sa

18 Oct 2026

Great paper from NVIDIA sharing an open midtraining recipe for agentic coding. NVIDIA kept SFT identical and added one midtraining stage. SW…

Great paper from NVIDIA sharing an open midtraining recipe for agentic coding. NVIDIA kept SFT identical and added one midtraining stage. SWE-bench Multilingual went from 55.8 to 62.8 on Nemotron 3.5

model-releasesdair-ai--x

17 Oct 2026

we’re getting used to it by now but DeepSeek calling this a “minor update” while moving Terminal-Bench 4.0 from 31.2 to 42.3 is very funny. …

model-releasesclem-delangue--x

we’re getting used to it by now but DeepSeek calling this a “minor update” while moving Terminal-Bench 4.0 from 31.2 to 42.3 is very funny. also reports 77.0 on DeepSWE (vs 74.0 for Opus 5 and 73.0 fo

17 Oct 2026

Why not run Qwen 3.8 Flash Next with the full n-gram-table?

model-releasesr-localllama

Already posted here https://www.reddit.com/r/LocalLLaMA/comments/1x85hfj/strata_qwen_38_flash_next_fast_but_whats_it/phavui9/ , but I believe this topic deserves it's own post. The PLE, aka the n-gram

17 Oct 2026

Anticipatory Commonsense with Moving Objects: An ASP-Based Framework for Predictive Reasoning with Interacting Trajectories

arXiv:2610.20867v1 Announce Type: new Abstract: We present a general declarative framework for anticipatory reasoning about interactive motion in complex dynamic environments. Our focus is on (explain

Behavioral Compositionality of Sparse Autoencoder Representations

arXiv:2610.21261v1 Announce Type: cross Abstract: Sparse autoencoders (SAEs) represent model activations with sparse, linearly decoded features, but the downstream effects of those features need not c

Beyond Context Length: Characterising the Difficulty of Long-Context Reasoning

arXiv:2610.22067v1 Announce Type: new Abstract: Long-context language models support ever longer context windows, and evaluations typically treat context length as the main indicator of difficulty. Ho

Can MLLMs Read the Room? A Multimodal Benchmark for Assessing Deception in Multi-Party Social Interactions

arXiv:2511.16221v2 Announce Type: replace-cross Abstract: Despite their advanced reasoning capabilities, state-of-the-art Multimodal Large Language Models (MLLMs) still lack a core component of human

CARE-SD: Explainable Reasoning Architecture for Local SLM Self-Disclosure Analysis in Textual Data

arXiv:2610.21151v1 Announce Type: cross Abstract: Cybersecurity interviews uncover critical security experiences, behaviors, and privacy concerns often not fully captured by conventional technical ass

Tutorials

EigDiff: Eigensample Diffusion - or how to significantly reduce training dataset size with marginal performance variation?

arXiv:2610.20999v1 Announce Type: cross Abstract: Diffusion models are notoriously expensive to train, requiring large-scale datasets to learn how to gradually reverse the noise from images over multi

tutorialsarxiv-cs-ai

Improving Environment-Agnostic Coding Agents: The Role of Scaffolding

arXiv:2610.22289v1 Announce Type: new Abstract: Stronger language models may increasingly construct useful procedures for themselves, reducing the advantage of developer-supplied scaffolding. We inves

Local Ai

InstantFusion: A Shared Latent Interface for Heterogeneous Image Generators

arXiv:2610.13944v2 Announce Type: replace Abstract: Diffusion/flow models have emerged as the de facto paradigms for image generation, giving rise to a diverse family of modern generators with increas

local-aiarxiv-cs-cv

Jev-as-a-Judge for RAG Claim Verification: Confidence-Gated Cascading Between a Non-Generative Verifier and a Frontier LLM

arXiv:2610.20634v1 Announce Type: new Abstract: Retrieval-augmented generation asks one question at volume: does this evidence support this claim? The answer is one bit, yet most pipelines obtain it f

Learning from Unverified Peers via Shared Residual Distillation

arXiv:2610.22080v1 Announce Type: new Abstract: A language model can generate several solutions to the same problem, some leading to correct answers and others to incorrect ones. Many post-training me

safetyarxiv-cs-ai

MBBQ: A Dataset for Cross-Lingual Comparison of Stereotypes in Generative LLMs

arXiv:2406.07243v4 Announce Type: replace Abstract: Generative large language models (LLMs) have been shown to exhibit harmful biases and stereotypes. While safety fine-tuning typically takes place in

MIDI-REPA: Improving Text-to-Music Generation with Symbolic Representation Alignment

model-releasesarxiv-cs-lg

arXiv:2610.20967v1 Announce Type: cross Abstract: State-of-the-art text-to-music models increasingly rely on scaling model capacity, data, and compute, while still learning musical structure directly

MSP: Modality Self-Play for Training Multimodal LLMs Without Paired Cross-Modal Data

arXiv:2610.20965v1 Announce Type: new Abstract: Large multimodal models have achieved remarkable progress by scaling model capacity and training on massive paired multimodal data. However, this paradi

safetyarxiv-cs-ai

Not X, but When: Diagnosing a Mandarin LLM Writing Tic through Contextual Appropriateness

arXiv:2610.22249v1 Announce Type: new Abstract: Recurring expressions in model-written text are often treated as signatures of machine authorship, yet most of them serve legitimate functions. Whether

researcharxiv-cs-cl

RLDF: A Region-Level Deepfake Dataset and Benchmark for Partial-Face Manipulation Detection

model-releasesarxiv-cs-cv

arXiv:2610.21438v1 Announce Type: new Abstract: Existing deepfake detection models are built upon manipu- lated images that are tampered with either exclusively fully swapped or fully synthesised face

The World Won't Wait: Evaluating and Teaching Agents Under a Live Clock

arXiv:2610.20665v1 Announce Type: new Abstract: Large language model (LLM) agents can solve complex interactive tasks, yet reasoning and tool calls take time. In continuously changing environments, an

researcharxiv-cs-ai

When Reality Outpaces Cognition: A New Scaling Principle for Reliable Intelligence

arXiv:2610.20764v1 Announce Type: new Abstract: This paper challenges the understanding capabilities of existing AI paradigms and emerging world models by asking a fundamental question: How quickly ca

4x R9700 builds with 127 used EPYCs: 10.7k all-in per machine. I built three (384GB VRAM)

model-releasesr-localllama

I saw the 4x R9700 build for a 10-person startup here a few days ago and wanted to share my results, since I opted to spend most of the money on the cards. I went with a super cheap used EPYC instead

ADAPT: Audio-Depth Action Planning with Transition-Aware Rewards for Lights-Out Industrial Navigation

arXiv:2610.15956v1 Announce Type: new Abstract: Vision-language-action models for embodied navigation typically rely on RGB observations, limiting their applicability in lights-out industrial environm

AesCode: Aesthetic Code Generation with Decoupled Cross-Modal Rewards

arXiv:2610.18860v1 Announce Type: new Abstract: Generating information-rich visual artifacts such as slides, posters, and dashboards as code is appealing because the outputs are structured, editable,

Are Benchmarks Reliable? Toward Structural Diagnosis via Item-Level Capability Boundaries

arXiv:2609.33196v2 Announce Type: replace Abstract: Evaluating large language models (LLMs) relies heavily on benchmark scores, yet aggregate metrics can obscure whether individual items serve their i

Are We Overthinking One-Step Generation?

arXiv:2610.20331v1 Announce Type: new Abstract: Diffusion and flow matching are now dominant frameworks for generative modeling in continuous domains, but they typically require multiple network evalu

researcharxiv-cs-lg

AtomSearch: Evidence-Grounded Tree Search for Agent Skill Optimization

arXiv:2610.19951v1 Announce Type: new Abstract: Agent skills adapt frozen models through editable external instructions. Revising several rules at once makes changes difficult to inspect, while judgin

b11658

model-releasesllama-cpp-releases

models : add maion-coder (#29778) Add Support to maion-coder arch iAdd Support to support Draft models Address comments Remove maion context shift function Co-authored-by: Karthik Vijayan kavijay@micr

Beyond Corrected Actions: Training LLM Agents on Textual Critiques

arXiv:2610.19367v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used for multi-turn agentic tasks that require interactions with users and environments over extended sequ

Beyond State Capacity: Why Retrieval Breaks in Mamba

arXiv:2610.19709v1 Announce Type: new Abstract: Mamba models underperform Transformers on in-context retrieval. This gap is usually attributed to Mamba's fixed-size recurrent state, which loses inform

researcharxiv-cs-cl

Confidently Wrong: Detecting Hallucinations in Financial Question Answering from LLM Internal States

arXiv:2607.11414v2 Announce Type: replace Abstract: Large language models (LLMs) in financial applications fail most consequentially when they are confidently wrong. Hedged, uncertain answers invite s

Cross-Lingual TTS for Indian Languages: A Benchmark and Inference-Time Duration Estimation Strategies

arXiv:2610.14826v1 Announce Type: cross Abstract: Recent alignment-free Text-to-Speech (TTS) models achieve high synthesis quality by removing explicit alignment modules and simplifying training. Howe

CSWBench: Evaluating Context-Sensitive Code-Switching in Creative Writing

arXiv:2610.20272v1 Announce Type: new Abstract: Large language models (LLMs) have demonstrated multilingual capabilities across diverse language generation and understanding tasks. Yet effective code-

CUSE: Fine-Grained Multimodal Unlearning with Counterfactual Guidance and On-Policy Self-Distillation

arXiv:2610.14763v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) can reproduce sensitive, copyrighted, or unsafe information from their training data. Effective unlearning must

safetyarxiv-cs-ai

Tutorials

Decoupled Marginal Sharpening for Training-Free Inference-Time Scaling

arXiv:2610.20483v1 Announce Type: new Abstract: We study how to improve LLM reasoning at inference time without updating model weights or using external reward models. Recent reward-free power-samplin

tutorialsarxiv-cs-lg

Do Theorem Provers Need Theorem-Proving Agents? mini-Lean-agent: A Coding-Agent Baseline for Lean

arXiv:2610.17331v1 Announce Type: new Abstract: General-purpose coding agents use standard shell and file interactions, yet formal theorem-proving agents often rely on specialized, ad-hoc tooling. Thi

Tutorials

Draft-Thinking: Learning Efficient Reasoning in Long Chain-of-Thought LLMs

arXiv:2603.00578v2 Announce Type: replace-cross Abstract: Long chain-of-thought~(CoT) reasoning improves the problem-solving capability of large reasoning models~(LRMs), but often incurs substantial i

tutorialsarxiv-cs-cl