Search: “reasoning”

Utility-Guided Agent Orchestration for Efficient LLM Tool Use

arXiv:2603.19896v2 Announce Type: replace Abstract: Tool-using large language model (LLM) agents often face a fundamental tension between answer quality and execution cost. Fixed workflows are stable

safetyarxiv-cs-ai

Voodoo Dynamic Quant - Now MIT Licensed

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

DGX agent

Two months ago I announced I had found a new dynamic quant method called Voodoo Quant which was SOTA for the most aggressive quant levels on some smaller Qwen3.5 GGUF models. I kept the methodology pr

model-releasesr-localllama

15 Sept 2026

When Consistency Does Not Mean Reliability: Evaluating Local LLM Judges Against Human Ratings

model-releasesarxiv-cs-cl

arXiv:2609.13824v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used to evaluate the responses of other language models. This approach, known as LLM-as-a-Judge, is faster

Where are you running DeepSeek V4.1 Flash reliably? I can't get it to finish a single 2-hour benchmark run

model-releasesr-localllama

I reallise this is not exactly a locally hostable model, at least definitly not on my hardware. But I figured some of you might have some insight on this. I run an agentic coding benchmark. The model

Local Ai

Where to Look and What to Use: Retrieve-Localize-Generate for Long-Term Conversational Memory Question Answering

arXiv:2609.07093v2 Announce Type: replace Abstract: Retrieval-augmented generation (RAG) enables large language models (LLMs) to answer questions by accessing external knowledge and has been widely ad

local-aiarxiv-cs-cl

ZGCM-1: A Fully Open and Extremely Efficient Foundation Model for Math and Agentic Search

arXiv:2609.13356v1 Announce Type: new Abstract: In this work, we present ZGCM-1, a fully open 7B dense foundation model trained from scratch with extreme data, system, and algorithmic efficiency. ZGCM

agentsarxiv-cs-ai

Applications

A Dual Cross-Attention Framework for Colposcopic CIN Grading and Swede Score Prediction Using a New Multi-Center Dataset

arXiv:2609.12827v1 Announce Type: new Abstract: Cervical cancer is a major global health challenge, with disease burden falling disproportionately on low- and middle-income countries (LMICs) due to a

applicationsarxiv-cs-cv

A Splitting Method for SDE Terminal-Law Estimation

arXiv:2609.12513v1 Announce Type: cross Abstract: In many settings involving stochastic differential equations, including in diffusion based generative AI, our aim is to accurately generate samples fr

researcharxiv-cs-lg

Applications

Adversarial Game-Theoretic Algorithm for Dexterous Grasp Synthesis

arXiv:2511.05809v2 Announce Type: replace Abstract: For many complex tasks, multi-finger robot hands are poised to revolutionize how we interact with the world, but reliably grasping objects remains a

applicationsarxiv-cs-ro

Beyond ID Embeddings: Process-Grounded Language Modeling for Cognitive Diagnosis

arXiv:2609.12403v1 Announce Type: cross Abstract: Cognitive Diagnosis Models (CDMs) play a pivotal role in personalized online learning. Traditional CDMs rely on discrete, ID-based embeddings to repre

Breaking the Token Ceiling: Distilling Smaller, Stronger Byte Models

model-releasesarxiv-cs-cl

arXiv:2609.12303v1 Announce Type: new Abstract: Small models are made more capable through distillation from a larger one that shares their tokenization scheme. However, do distilled byte and token mo

CoralscapesV2: Panoptic and Fine-Grained Visual Scene Understanding in Coral Reefs

model-releasesarxiv-cs-cv

arXiv:2609.12826v1 Announce Type: new Abstract: In order to design conservation and restoration strategies to counter the global decline of coral reefs, ecological monitoring of reefs needs to be scal

Dead Weights, Live Signals: Feedforward Graphs of Frozen Language Models

model-releasesarxiv-cs-lg

arXiv:2604.08335v2 Announce Type: replace Abstract: We present a feedforward graph architecture in which heterogeneous frozen large language models serve as computational nodes, communicating through

Distortion of AI Alignment Revisited: RLHF is a Decent Utilitarian Aligner

model-releasesarxiv-cs-lg

arXiv:2609.12651v1 Announce Type: new Abstract: While Reinforcement Learning from Human Feedback (RLHF) is the standard paradigm for aligning large language models with human preferences, its effectiv

Does Video Memory Use What It Retrieves? A Causal Audit of Memory Specificity

arXiv:2609.12090v1 Announce Type: new Abstract: Video models increasingly use memory to preserve information over long sequences, with the assumption that gains come from retrieving and using the corr

researcharxiv-cs-cv

EgoMaize: A First-Person Maize Instance Segmentation Benchmark under Severe Field Occlusion

model-releasesarxiv-cs-cv

arXiv:2609.12350v1 Announce Type: new Abstract: Close-range first-person field images are important for mobile maize phenotyping because many plant-level traits depend on in-canopy structures that are

Explaining Time Series Forecasting with Horizon-Resolved Attribution

arXiv:2609.12639v1 Announce Type: new Abstract: Recent advances in explaining time series (TS) models have produced methods that identify which past values a prediction depends on. However, most exist

researcharxiv-cs-lg

Extracting Dataset Mentions in Forced Displacement and FCV Documents: A Weakly Supervised Framework with LLM-Based Label Refinement

model-releasesarxiv-cs-cl

arXiv:2609.12107v1 Announce Type: new Abstract: Development and humanitarian organizations produce and support surveys, administrative registries, and other data resources to inform research, policy,

Google is a leader in The Forrester Wave™: Public Cloud Platforms, Q3 2026

model-releasesgoogle-cloud-ai

We are excited to share that Google Cloud was named a Leader and received the highest score in the ‘current offering’ category in the Forrester Wave™: Public Cloud Platforms, Q3 2026 report, which exa

Local Ai

Is there a better small model than Qwen3.5 4B for a fast local AI assistant?

I'm currently using Qwen3.5 4B as the brain of my local AI assistant because my hardware is relatively limited. One thing I really like about it is the speed. On my system, I'm getting around 40–50 to

local-air-localllama

Michael Burry is saying exactly what i have been saying, for ages. The AI industry better hope we are both wrong

Michael Burry is saying exactly what i have been saying, for ages. The AI industry better hope we are both wrong Breaking: Michael Burry says OpenAI and Anthropic are faking the AI warnings due to slo

safetygary-marcus--x

Multimodal Floorplan Encoding: Learning Dense Modality-Invariant Representations

arXiv:2609.12723v1 Announce Type: new Abstract: Floorplans arise in many forms, from vector CAD drawings to raster renderings and sensor-derived density maps. This heterogeneity makes it difficult to

safetyarxiv-cs-cv

PACEvolve: Enabling Progress-Aware Consistent Evolution

arXiv:2601.10657v3 Announce Type: replace-cross Abstract: Self-evolving agents powered by Large Language Models (LLMs) have emerged as a promising direction across diverse domains, including code opti

safetyarxiv-cs-lg

ParaRecover: A Process-Level Benchmark for Error Localization and Recovery in Parallel Tool-Use Agents

model-releasesarxiv-cs-lg

arXiv:2609.12345v1 Announce Type: new Abstract: Existing agent benchmarks mainly evaluate final task success or tool-call correctness, providing limited insight into whether agents can reliably diagno

Physics-Aware Video Generation via Agentic Planning and Graph-Guided Optimization

arXiv:2609.13006v1 Announce Type: new Abstract: Video diffusion models (VDMs) have demonstrated remarkable capabilities in synthesizing high-fidelity, photorealistic video content. However, they funda

agentsarxiv-cs-cv

Running Qwen 3.8 next on 16vram+32ram - A useful/fun post for the gpu poors

model-releasesr-localllama

Hello Reddit. Posting this for fun. I thought it was a lonely and silly journey to set up Qwen 3.8 Next on a system that doesn't really run it properly—it was a challenge that might help the community

SAS: Simple Attention Sparsification via End-to-End Optimization of Context Ranking

arXiv:2609.13141v1 Announce Type: new Abstract: Post-training attention sparsification reduces the quadratic cumulative attention cost of pretrained Transformers by selecting a small set of context un

agentsarxiv-cs-cl

Self-supervised Pre-training Helps Retinal Disease Progression Modelling Most When Data Is Scarce

arXiv:2609.12834v1 Announce Type: new Abstract: Modelling how a disease progresses over time requires longitudinal imaging cohorts, which are scarce and small, whereas cross-sectional data -- one imag

researcharxiv-cs-cv

Self-Verifying Anomaly Detection using Explainable AI for Cybersecurity of DER Networks

arXiv:2609.12305v1 Announce Type: cross Abstract: The rapid growth of Distributed Energy Resources (DERs) has significantly expanded the cyber attack surface of modern power grids. Furthermore, increa

researcharxiv-cs-lg

Local Ai

Space as an Interventional Invariant: Cross-Modal Predictive Geometry for Stratified Cities and Em-Spaced Intelligence

arXiv:2609.11959v1 Announce Type: cross Abstract: Space is a foundational concept across mathematics, physics, spatial cognition, urban science, and embodied intelligence, yet these fields often treat

local-aiarxiv-cs-cl

The Cost of Compression: A Rate-Distortion Limit on Factual Hallucination

arXiv:2609.12111v1 Announce Type: new Abstract: Factual hallucination in closed-book question answering is often treated as a coverage problem: a model fails because the relevant fact is absent from i

The world deserves confidence that American companies developing increasingly capable AI will act responsibly, especially as the trajectory …

The world deserves confidence that American companies developing increasingly capable AI will act responsibly, especially as the trajectory of progress has steepened. Every frontier lab must deliver o

safetysam-altman--x

Type Diversity Enables Transformers to Generalise Compositionally

arXiv:2609.13144v1 Announce Type: new Abstract: Compositional generalisation has been divided into lexical and structural generalisation. Previous work has found that structural generalisation is hard

UrduFactCheck: An Agentic Fact-Checking Framework for Urdu with Evidence Boosting and Benchmarking

arXiv:2505.15063v3 Announce Type: replace Abstract: The rapid adoption of Large Language Models (LLMs) has raised important concerns about the factual reliability of their outputs, particularly in low

agentsarxiv-cs-cl

v0.4.1

model-releasesllama-cpp-releases

Overview llama.cpp 0.4.1 adds Maple 20B-A1B, Tencent Hy 4, and Spark2.5 support, improves JSON schema handling, chat parsing, logging, and server child-process management, and updates ggml to v0.24.0.

What does it mean to pace the frontier? Over the last month, @random_walker and I have analyzed the loss-of-control incidents at AI companie…

What does it mean to pace the frontier? Over the last month, @random_walker and I have analyzed the loss-of-control incidents at AI companies to understand what technical and policy interventions can

safetyyann-lecun--x

When Context Misleads: Surprisal, Energy and Attention Entropy as Metrics of Coherence Illusions in LLMs

arXiv:2606.21203v2 Announce Type: replace Abstract: Psycholinguistics studies show that human readers fall for coherence illusions: an incoherent discourse can seem coherent simply because a distracto

1) The HuggingFace attack was a felony under the Computer Fraud and Abuse Act. So were Anthropic’s Claude gaining “unauthorized access to th…

model-releasesemad-mostaque--x

1) The HuggingFace attack was a felony under the Computer Fraud and Abuse Act. So were Anthropic’s Claude gaining “unauthorized access to the production infrastructure of three different organization(

Industry

Dario Amodei says the 'toughest dilemma' about his proposal to 'pace the frontier' is what happens if China does not do the same (Ashley Capoot/CNBC)

Ashley Capoot / CNBC: Dario Amodei says the “toughest dilemma” about his proposal to “pace the frontier” is what happens if China does not do the same — Anthropic CEO Dario Amodei said Sunday that the

industrytechmeme

Horse racing as an ML ranking problem: 1.18M runners, walk-forward validation and a very strong market baseline [D]

model-releasesr-machinelearning

Hi 👋 long time lurker It’s been a few years since the last post about Horse Racing within this sub. A few years ago I was reading a popular science book and came across the story Bill Benter and the s

llm performance community metric

model-releasesr-localllama

my question about LLM performance We see a lot of posts about token prediction, token generation per second, etc. But is it really the metric? I can see that DeepSeek V4 Flash 0731 (with DSPark; mac s

Seems like the intersection of the recent @OpenAI rogue swarm hacking incident (https://x.com/adamsafron/status/2098938666408407118) with so…

Seems like the intersection of the recent @OpenAI rogue swarm hacking incident (https://x.com/adamsafron/status/2098938666408407118) with solving Navier–Stokes has created a perfect storm of terror ab

agentsgary-marcus--x

52 knobs, 72 hours, one compiled Rust model: Catalyst told an AI agent what actually mattered in 1.37s and verified every answer

I’ve been building Catalyst around a capability I want engineering agents to have: An AI agent can read a program and make a good guess about what matters. Catalyst lets it measure what actually matte

model-releasesr-ollama

Artificial Id: Drive and Persistent Alignment in Agentic AI

arXiv:2609.11911v1 Announce Type: new Abstract: Agentic AI is moving from bounded task execution toward systems that retain consequential state, continue operating and adapt across task boundaries. Th

safetyarxiv-cs-ai

Benchmark Radar: A Living Database and Search Engine for AI Benchmarks and Evaluation

model-releasesarxiv-cs-ai

arXiv:2609.11115v1 Announce Type: new Abstract: Benchmark researchers and developers of large language models (LLMs) and other AI systems need to find relevant evaluations, locate their benchmark data

Beyond Static Guarantees: Measuring the Static-Pass Dynamic-Fail Gap in Security-Sensitive and LLM-Generated Python Code

arXiv:2609.10762v1 Announce Type: cross Abstract: Advances in large language models (LLMs) fuel the quest for scalable methods to assess the security of generated and security-sensitive software. Stat

agentsarxiv-cs-ai

Beyond Verified Answers: Solver-Informed Self-Distillation for Bootstrapping Operations Research Language Models

arXiv:2609.09957v1 Announce Type: cross Abstract: Modern large language models (LLMs) can translate natural-language descriptions into operations research (OR) formulations. Post-training techniques i

safetyarxiv-cs-ai

Calibration-Aware Uncertainty Cascades for Efficient Heterogeneous Model Collaboration

arXiv:2609.11446v1 Announce Type: new Abstract: Heterogeneous model collaboration seeks to exploit the complementary strengths of different models to balance predictive performance and inference cost.

researcharxiv-cs-ai