Search: “models”

A multicenter benchmark and clinically structured metric for coronary CTA report generation

arXiv:2609.00909v1 Announce Type: new Abstract: Reliable evaluation of automated coronary computed tomography angiography (CCTA) report generation requires standardized multicentre benchmarks and clin

Human

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Model Releases

AutoXRD: Autonomous LLM Agents and Comprehensive Evaluation for Powder Diffraction Analysis

DGX agent

arXiv:2609.00070v1 Announce Type: cross Abstract: Powder X-ray diffraction (XRD) is central to materials characterization, yet reliable end-to-end automation remains challenging. An XRD agent must int

model-releasesarxiv-cs-ai

2 Sept 2026

Behaviorally Grounded User Profiles from the Wild for Personalized Alignment and Multi-Perspective Reasoning

arXiv:2609.00014v1 Announce Type: cross Abstract: Persona-driven techniques increasingly adapt large language models (LLMs) to diverse contexts. However, existing methods predominantly rely on rigid,

safetyarxiv-cs-ai

Beyond Blind Compliance: Benchmarking Task Verification in OCR Reasoning

arXiv:2609.00232v1 Announce Type: new Abstract: Multimodal Large Language Models (MLLMs) have achieved strong performance on OCR-centric document understanding and text-rich visual reasoning benchmark

Local Ai

BiasGym: A Simple and Generalizable Framework for Analyzing and Removing Biases through Injection

arXiv:2508.08855v5 Announce Type: replace-cross Abstract: Understanding biases and stereotypes encoded in the weights of Large Language Models (LLMs) is crucial for developing effective mitigation str

local-aiarxiv-cs-ai

Can LLMs Design Video Coding Tools? A Case Study on Planar Mode

arXiv:2609.01535v1 Announce Type: cross Abstract: This paper explores whether large language models (LLMs) can design video coding tools, a highly challenging task due to the intricate algorithmic cou

CoMMET: A Psychologically Grounded Benchmark for Evaluating Theory of Mind in Multimodal LLMs

arXiv:2603.11915v2 Announce Type: replace Abstract: Theory of Mind (ToM)-the ability to reason about the mental states of oneself and others-is a cornerstone of human social intelligence. As Multimoda

Compile, Don't Memorize: A Context Compilation Architecture (CCA) for In-Context Learning

arXiv:2609.00759v1 Announce Type: new Abstract: Large language models (LLMs) increasingly handle in-context learning (ICL) tasks where a long, novel context defines the rules, knowledge, and output sc

agentsarxiv-cs-cl

Confirmed bolting Q8 NGram into IQ4 Qwen no speed degradation

model-releasesr-localllama

This came from another thread or comment. I forgot exactly where, but the basic idea was to replace the 51B N-gram layer in Qwen 3.8 Next with a much higher precision version. Someone running a 5090 r

Controllable Image Captioning with Prompt-Conditioned Scene Rewards

arXiv:2609.00709v1 Announce Type: cross Abstract: Large Vision-Language Models produce fluent image descriptions but offer limited semantic control: users cannot reliably specify whether captions shou

DSG: Dynamic 3D Scene Graph Construction for Embodied Agents in Changing Indoor Environments

model-releasesarxiv-cs-ro

arXiv:2609.00619v1 Announce Type: new Abstract: In indoor environments, object positions frequently change due to human activities or embodied-agent interactions, causing previously constructed scene

Applications

Efficiently Estimating Optimal Hyperparameter Scaling Laws through Power-Law Entropy Search

arXiv:2609.01431v1 Announce Type: cross Abstract: Optimal hyperparameter scaling laws describe how the best hyperparameters for large language model (LLM) training change with model and data scale, en

applicationsarxiv-cs-ai

EmbodiedSkills: A Unified Framework for Orchestrating, Training, and Deploying VLA Agents

arXiv:2609.01281v1 Announce Type: cross Abstract: Vision-language-action (VLA) models map visual observations and language instructions directly to robot actions, but long-horizon tasks require more t

agentsarxiv-cs-ai

ExBind: A Controlled Diagnostic Benchmark for Visual-to-Executable Correspondence

arXiv:2609.01344v1 Announce Type: new Abstract: Multimodal coding and editing systems must map a visible or semantic referent to the exact executable object that can be edited. A wrong reference may s

Exploring Collaboration between a language and a non-language agent

arXiv:2609.00474v1 Announce Type: cross Abstract: LLMs are increasingly deployed as orchestrators that coordinate specialized subagents to solve complex tasks through natural language. However, in man

Flawed in Nature, Perfect through Evolution

arXiv:2609.00129v1 Announce Type: cross Abstract: The performance of artificial intelligence (AI) and machine learning (ML) models degrades when the problem they were trained on drifts. This is a near

agentsarxiv-cs-ai

FlexP-SFT: A Flexible Aggregation-Free Framework for On-Device Personalized Split Federated Fine-Tuning of LLMs

arXiv:2508.10349v2 Announce Type: replace-cross Abstract: To fine-tune large language models (LLMs) over private data, federated learning (FL) has emerged as a promising paradigm. However, the prohibi

Google launches Gemini 3.8 Flash Cyber for partners in its new Fairwind Program and says Gemini 3.8 Flash beats Claude Opus 5 and GPT-5.6 Sol on some benchmarks (Google)

Google: Google launches Gemini 3.8 Flash Cyber for partners in its new Fairwind Program and says Gemini 3.8 Flash beats Claude Opus 5 and GPT-5.6 Sol on some benchmarks — Our newest Gemini models deli

model-releasestechmeme

Hints Help But Do They Teach? Evaluating Skills Transfer in Code Generation

arXiv:2609.01106v1 Announce Type: cross Abstract: When a hint turns a failing generated program into a passing one, does it provide missing information or merely steer the model toward a solution it c

HiveTraceGuard-Pro: A Compact Generative Guardrail for Prompt Injection, Jailbreaks, and Adversarial Obfuscation

arXiv:2609.01046v1 Announce Type: cross Abstract: Production LLMs must handle inputs that attempt to override system instructions, bypass safety policies or elicit harmful responses. A common mitigati

Knowledge Distillation During Mid-Training Favors Reasoning over Factual Recall

arXiv:2609.01532v1 Announce Type: new Abstract: Logit-based knowledge distillation (KD) is used to train smaller language models (LMs) via supervision from stronger teachers, but whether its benefits

researcharxiv-cs-cl

Lingua Franca or Probing Artifact? Rethinking Latent Language in Multilingual LLMs

arXiv:2609.00155v1 Announce Type: cross Abstract: Latent language identification is often used to argue that multilingual language models route computation through language-specific states, such as En

researcharxiv-cs-ai

MiNER: Fine-Tuned Biomedical Natural Language Processing for Malaria Disease Entity Recognition in Clinical Texts

arXiv:2609.00073v1 Announce Type: new Abstract: Malaria remains a significant global health burden, necessitating continuous research efforts to understand its complex molecular mechanisms, epidemiolo

researcharxiv-cs-ai

Multi-Head Self Attention is a Parameter Identification Mechanism

arXiv:2609.01231v1 Announce Type: new Abstract: We prove that a multi-head scaled dot product attention can be viewed as a parameter identification strategy. The ratio of unidentified parameters to th

mzCache: On-Device LLM Memory Management under Multitasking

arXiv:2609.01338v1 Announce Type: cross Abstract: On-device mobile Large Language Model (LLM) inference is gaining significant attention. However, mobile devices operate in highly dynamic multitasking

On the Existence of Consistent Adversarial Attacks in High-Dimensional Linear Classification

arXiv:2506.12454v2 Announce Type: replace-cross Abstract: What fundamentally distinguishes an adversarial attack from a misclassification due to limited model expressivity or finite data? In this work

researcharxiv-cs-lg

PersuaRL: Reinforcement Learning-Driven Multi-Expert Selection for Persuasive Dialogue Generation in Insurance

arXiv:2609.01188v1 Announce Type: new Abstract: Large Language Models (LLMs) are revolutionizing digital communication by powering conversational agents deployed across domains such as customer servic

Polish ModernBERT: The Long and Short of Polish Language Understanding

arXiv:2609.01379v1 Announce Type: new Abstract: Encoder-only Transformers remain effective for discriminative and representation-learning tasks, yet Polish encoders still largely rely on BERT/RoBERTa-

Replicating TRACE: A Practitioner's Guide to Its Threshold and Particle Budget

arXiv:2609.01108v1 Announce Type: new Abstract: TRACE (Math & Lienhart, arXiv:2602.01135) reads causal graphs over event types out of a pretrained autoregressive sequence model by thresholding a per-p

Revisiting Cross-View Completion: Self-Supervised Pre-Training via Reconstruction Error Comparison

arXiv:2609.01530v1 Announce Type: new Abstract: Self-supervised pre-training via cross-view completion learns strong features for 3D vision from co-visible regions of image pairs. However, the referen

RPCBench: A Benchmark for Proactive Premise Critique in LLM-based Recommendation

arXiv:2609.00918v1 Announce Type: new Abstract: Large language models are increasingly used as interactive recommender assistants. Their evaluation should therefore go beyond plausible item recommenda

Safin-1: Safety from Within through Memory-Native State Evolution

arXiv:2609.00092v1 Announce Type: new Abstract: Long-horizon complex tasks require foundation models to accumulate information, maintain internal states, and adapt over extended interactions. Safety s

safetyarxiv-cs-lg

TempCloze: Can Video-LLMs Identify the Missing Middle?

arXiv:2609.01515v1 Announce Type: cross Abstract: Temporal reasoning benchmarks for Video-LLMs are often mediated by language, leaving room for linguistic shortcuts from option wording, answer correla

The 1967 Spider-Man TV Show intro, updated to live action with MiniMax H3

model-releasesr-stablediffusion

R2V Rendered at 0.9 MP (1280x736 then upscaled using RTX (Ultra) to 1920x1080. Edited and merged using OpenShot video editor. This was all run on my Windows 11 machine, RTX 4060 ti (16 GB) and 64 GB R

The Assistant's Ideal Self

arXiv:2609.00304v1 Announce Type: new Abstract: Models express values and welfare-relevant self-reports, but it is unclear whether these outputs reflect stable preferences or a stable self. We thus in

safetyarxiv-cs-ai

Tutorials

The Structure of Quantization Damage in LLMs: Why the Next Bit Should Be Spent Globally

arXiv:2609.01587v1 Announce Type: cross Abstract: Post-training quantization (PTQ) is widely used to reduce the cost of serving large language models (LLMs), but its accuracy cost is uneven and is oft

tutorialsarxiv-cs-cl

Think Like a Doctor: Conversational Diagnosis through the Exploration of Diagnostic Knowledge Graphs

arXiv:2602.01995v2 Announce Type: replace Abstract: Conversational diagnosis requires multi-turn history-taking, where an agent asks clarifying questions to refine differential diagnoses under incompl

Trust Your Guide Only When Certain: Uncertainty-Aware Sparse Alignment at Inference Time

arXiv:2609.00624v1 Announce Type: new Abstract: A prominent paradigm in inference-time alignment employs lightweight supervisors to steer Large Language Models (LLMs). Through empirical analysis, we i

safetyarxiv-cs-cl

UI-Venus-2 Technical Report

arXiv:2609.00028v1 Announce Type: new Abstract: Multimodal GUI agents have emerged as a promising paradigm for digital task automation, yet transitioning from benchmark-oriented models to dependable r

WHALE: A Simple Recipe for Joint Harness-Weight Optimization

arXiv:2609.00196v1 Announce Type: cross Abstract: Agent performance depends jointly on the model parameters and the executable harness code that manages context and control flow. Optimizing either com

agentsarxiv-cs-ai

Which one will you choose and why, between R9700 32GB vs W7800 48GB?

model-releasesr-localllama

I'm planning to upgrade my workstation (linux with 5700X/64GB DDR4) for local inference and pytorch training. I'm trying to decide between: 2× AMD Radeon AI PRO R9700 32GB 2× AMD Radeon PRO W7800 48GB

Your favorite fastest abliterated/safety removed 3.6 and 3.8 27b?

model-releasesr-localllama

Not written by AI all mistakes mine. I saw people on the subreddit saying that 3.6 works better without thinking. It made me want to know for certain about which is better, 3.6 or 3.8 for low thinking

A Hybrid PEM-GP Framework for Uncertainty-Aware System Identification of Quadcopters

arXiv:2608.30433v1 Announce Type: new Abstract: Accurate dynamic models play a central role in achieving reliable control of quadcopters. Classical system identification methods remain widely used, ma

researcharxiv-cs-ro

An Agentic Retrobiosynthesis Framework with Learned Frontier Selection

arXiv:2608.30702v1 Announce Type: cross Abstract: Large language models are increasingly used as agents for multistep retrosynthesis, raising the question of how much their search policy contributes i

An Efficient Sparse Fine-Tuning with Low Quantization Error via Neural Network Pruning

arXiv:2502.11439v3 Announce Type: replace-cross Abstract: Fine-tuning is an important step in adapting foundation models such as large language models to downstream tasks. To make this step more acces

researcharxiv-cs-ai

An Open-Source, Event-Driven Pipeline for Cryptocurrency Market Data: Ingestion, Forecasting, and On-Chain Fraud Detection

arXiv:2608.29973v1 Announce Type: new Abstract: Cryptocurrency markets generate high-frequency, multi-source data that is expensive to work with unless a team already has commercial-grade streaming an

APIFlow-Bench: Measuring Whether Agents Survive Long, Dependent API Workflows

arXiv:2608.29128v1 Announce Type: new Abstract: Tool-using agents are commonly evaluated by a single bit: whether an end-to-end workflow completed. This metric fails to distinguish failures that matte

Tutorials

ARMOR: Manifold-Oriented Training for Adversarially Robust Aerial Object Detection under Data Scarcity

arXiv:2608.29510v1 Announce Type: new Abstract: Aerial object detection is increasingly deployed in real-world applications, but models remain vulnerable to physical, universal adversarial patches tha

tutorialsarxiv-cs-cv