Search: “models”

Locking Pretrained Weights via Deep Low-Rank Residual Distillation

researchapple-ml-research

The quality of open-weight language models has dramatically improved in recent years. Sharing weights greatly facilitates model adoption by enabling their use across diverse hardware and software plat

Model Releases

OmniEdit-Bench: A Comprehensive Benchmark for Instruction-based Video Editing

DGX agent

arXiv:2608.05049v1 Announce Type: new Abstract: Instruction-based video editing (IVE) is an emerging field with broad applications, yet evaluating editing models remains challenging. Existing benchmar

model-releasesarxiv-cs-cv

6 Aug 2026

Reading Between the Frames: Interpreting Implicit and Non-literal Meaning in Social Media Videos

arXiv:2608.04939v1 Announce Type: new Abstract: Social media videos often communicate meanings that go beyond their visible actions, captions, or speech. A mundane clip may become humorous, ironic, or

RepoProbe: Benchmarking Architecture-Aware Repository Comprehension with Checklists

arXiv:2608.04783v1 Announce Type: cross Abstract: The integration of Large Language Models (LLMs) into software engineering has shifted the focus from function-level generation to repository-scale ass

RooflineBench: A Benchmarking Framework for On-Device LLMs via Roofline Analysis

arXiv:2602.11506v4 Announce Type: replace-cross Abstract: The transition toward localized intelligence through Small Language Models (SLMs) has intensified the need for rigorous performance characteri

SEAR: Simple and Efficient Adaptation of Visual Geometric Transformers for Unpaired RGB+Thermal 3D Reconstruction

arXiv:2603.18774v2 Announce Type: replace Abstract: Foundational feed-forward visual geometry models enable accurate and efficient camera pose estimation and scene reconstruction by learning strong sc

Social Pressure Breaks Majority Voting in LLM Safety Panels

arXiv:2608.04415v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used to detect unsafe content. A common approach is to combine judgments from a panel of models to correct

safetyarxiv-cs-cl

Teaching Nemotron Greek: Mining a Corpus, Adapting Retrieval, and Grounding Generation for Modern Greek across Specialist Domains

arXiv:2608.05138v1 Announce Type: cross Abstract: Modern Greek is absent from NVIDIA's Nemotron retrieval models and from major multilingual retrieval benchmarks, despite being important for retrieval

Text2GraphQuery-Bench: A Text to Graph Query Benchmark

arXiv:2602.11745v2 Announce Type: replace Abstract: Graph models are fundamental to data analysis in domains rich with complex relationships. Unlike SQL, which benefits from a rel- atively unified sta

When Prompts Become Pixels: Prompt-Region Grounding for Multimodal Reasoning

arXiv:2608.04726v1 Announce Type: new Abstract: Multimodal large language models increasingly reason over screenshots and documents where the task itself may be written in pixels. Yet benchmarks usual

AI Security Leaderboard: Methodology, Results and Minimal Standard

arXiv:2608.03070v1 Announce Type: cross Abstract: Frontier AI model developers increasingly rely on layered safeguards to prevent catastrophic misuse, but little public evidence exists on how much pro

ChartAnno: Evaluating MLLMs for Chart Annotation Generation

arXiv:2608.03464v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) have made significant progress in chart understanding, generation, and editing, but their ability to annotate e

DenialRAG: Single-Document RAG Poisoning via Embedded Parametric Denial

arXiv:2608.02678v1 Announce Type: cross Abstract: Retrieval-augmented generation (RAG) systems are vulnerable to corpus poisoning: an attacker who inserts a crafted document into the retrieval corpus

EduClaw-Bench: A Long-Horizon Benchmark for Pedagogical LLM Agents with Simulated Learners

arXiv:2608.03206v1 Announce Type: cross Abstract: Large language models (LLMs) power educational applications from tutoring to essay scoring, but each is a point solution to a single task, and only re

FinVerse: Financial Time-Series Benchmark

arXiv:2608.03259v1 Announce Type: cross Abstract: As time-series foundation models have emerged, the need for benchmarks that can evaluate their forecasting ability in meaningful ways has become incre

Local Ai

I took a local OCR model's accuracy from 60% to 99%

I built a local OCR pipeline a few days ago, and it turned into a surprisingly interesting experiment—taking accuracy from around 60% to 99%. I wrote a short blog about what worked, what failed, and t

local-air-ollama

Improved Quantum Algorithms for Reinforcement Learning Under a Generative Model

arXiv:2608.02826v1 Announce Type: cross Abstract: Reinforcement learning is a subfield of machine learning that studies how an agent interacts with an environment in order to extract as large a reward

safetyarxiv-cs-ai

Agents

LACE: Large Language Model Aided Multi-Agent Framework for Agile RISC-V Instruction Extension

arXiv:2608.02915v1 Announce Type: cross Abstract: Domain-specific Instruction Set Architecture eXtensions (ISAX) are widely adopted in the RISC-V ecosystem to accelerate emerging workloads, but implem

agentsarxiv-cs-cl

LFM2.5-2.6B on a OnePlus 13 at 17 tok/s ~ Pure CPU

model-releasesr-localllama

As you all know the model is 2.69B parameters with a 128K context window and purpose-built for multi-step agent workflows. What you are seeing is the Q4_K_M GGUF running on my own inference engine bui

Applications

MaterialFusion: High-Quality, Zero-Shot, and Controllable Material Transfer with Diffusion Models

arXiv:2502.06606v3 Announce Type: replace Abstract: Manipulating the material appearance of objects in images is critical for applications like augmented reality, virtual prototyping, and digital cont

applicationsarxiv-cs-cv

PASE: Leveraging the Phonological Prior of WavLM for Low-Hallucination Generative Speech Enhancement

arXiv:2511.13300v1 Announce Type: cross Abstract: Generative models have shown remarkable performance in speech enhancement (SE), achieving superior perceptual quality over traditional discriminative

researcharxiv-cs-ai

Route-Align-Verify for Functional Correctness in Code Generation

arXiv:2608.03341v1 Announce Type: cross Abstract: Large language models (LLMs) have substantially improved code generation, yet achieving strong functional correctness remains difficult, especially fo

Single Canonical Prompts Underestimate LLM Safety's Surface-Form Sensitivity

arXiv:2608.02665v1 Announce Type: cross Abstract: A benchmark score is a measurement instrument, yet most benchmarks read each item at a single canonical surface form. We ask whether that reading is f

SPADE: An Input-Adaptive Sparse Attention Engine for Fast Video Diffusion Models Inference

arXiv:2608.03335v1 Announce Type: new Abstract: Video diffusion transformers (vDiTs) generate high quality but pay quadratic self-attention cost, making inference prohibitive at video-token scales. Th

safetyarxiv-cs-cv

Local Ai

The Tell-Tale Trace: Detecting Reasoning Failures in LLMs Using Chain-of-Thought Dynamics

arXiv:2608.03291v1 Announce Type: cross Abstract: Chain-of-thought (CoT) reasoning improves large language model (LLM) performance while also providing an observable interface to the model's reasoning

local-aiarxiv-cs-ai

UHP Detection: LVLMs have their Unique Hallucination Pattern in the Consistency Space

arXiv:2608.03817v1 Announce Type: cross Abstract: Large vision--language models (LVLMs) demonstrate strong multimodal reasoning capabilities but remain prone to hallucination, where model predictions

researcharxiv-cs-ai

Wiring Beats Blending: What Transfers Between Transformer Sizes -- and What Doesn't

model-releasesarxiv-cs-lg

arXiv:2608.02829v1 Announce Type: new Abstract: Model families train every size from scratch. Can a pretrained large model be converted into a smaller sibling? We characterize the 1.4B->410M conversio

A Benchmark Dataset for MLLM-Generated Image Detection: GPT Image2 & Nano Banana2

arXiv:2608.01258v1 Announce Type: new Abstract: The realism of images generated by multimodal large language models (MLLMs), such as GPT Image2 and Nano Banana2, has improved rapidly in recent years.

AdaMTP: An Adaptive Training Paradigm for Multi-Token Prediction

arXiv:2608.00434v1 Announce Type: new Abstract: Multi-Token Prediction (MTP) has emerged as an effective paradigm that augments a shared Large Language Model backbone with auxiliary heads, training th

Advancing Relevance Measurement with Vision-Language Models for Web-Scale Search

arXiv:2608.02446v1 Announce Type: cross Abstract: Relevance evaluation plays a crucial role in personalized search systems, serving as a guardrail alongside user engagement metrics to ensure that sear

safetyarxiv-cs-lg

Analyzing Speech Condition Effects in Dysarthric ASR: A Layer-wise Probing Study

arXiv:2608.01865v1 Announce Type: new Abstract: Automatic speech recognition (ASR) performance degrades sharply on dysarthric speech, yet how disordered articulation reshapes a model's internal repres

Local Ai

Beyond Global Latents: Chunk-Based Sparse Grid VAE for Scalable 3D Modeling

arXiv:2608.02016v1 Announce Type: new Abstract: Sparse voxel grids preserve the spatial structure needed for detailed 3D reconstruction, but their memory still grows rapidly with resolution as active

local-aiarxiv-cs-cv

CADENA: Stepwise CAD Reverse Engineering

arXiv:2608.00799v1 Announce Type: new Abstract: Computer-Aided Design (CAD) underpins modern engineering, yet converting existing shapes into editable models still demands substantial expert effort. M

Can Humans Dream of Electric Sheep? Human-Written Samples for Fine-Grained Vision-and-Language Hallucination Benchmarking

arXiv:2608.01021v1 Announce Type: cross Abstract: In an age of rapid model turnover, how do we make hallucination evaluation more perennial? We explore whether human-written hallucination samples coul

researcharxiv-cs-cl

Can Language Models Identify Shadow Trading Targets? An NLP Evaluation of SEC Enforcement Theory

arXiv:2608.01322v1 Announce Type: new Abstract: Shadow trading -- trading in a peer firm's securities on the basis of material nonpublic information (MNPI) about an 'economically linked' company -- is

researcharxiv-cs-cl

Don't Judge a Book by its Cover: Testing LLMs' Robustness Under Logical Obfuscation

arXiv:2602.01132v2 Announce Type: replace Abstract: Tasks such as solving arithmetic equations, evaluating truth tables, and completing syllogisms are handled well by large language models (LLMs) in t

EgoIntent: A Pre-Outcome Micro-Step Benchmark for Understanding What, Why, and Next

arXiv:2603.12147v2 Announce Type: replace Abstract: Egocentric video provides a natural modality for studying human behavior, but conventional visual understanding captures mainly observable scenes, o

FinHardBench: Can LLMs Generate Latency-Aware Hardware for Financial Computing?

arXiv:2608.00909v1 Announce Type: new Abstract: Can large language models generate not just correct, but fast hardware? This paper investigates the question in financial FPGA design, where 5-10 nanose

From Direction to Magnitude: How Multimodal Instruction-Tuning Reorganizes the Geometric Encoding of Identity-Specifying Prompts in Transformer Hidden States

arXiv:2607.09842v2 Announce Type: replace-cross Abstract: We investigate whether identity-specifying system prompts produce statistically distinguishable geometric fingerprints in the hidden-state tra

Hardware

Generate Trajectories, Reasoning Traces, and Auto-Labels with NVIDIA Alpamayo 2 Super

NVIDIA Alpamayo 2 Super is a publicly available 34‑billion‑parameter vision–language–action model that merges a 32‑B Cosmos 3 Super Reasoner with a 2‑B action‑expert diffusion network. It produces uni

hardwarenvidia-developer

GIFT: Geometry-Invariant Fine-Tuning for Non-Lambertian Monocular Depth Estimation

arXiv:2608.02068v1 Announce Type: new Abstract: Monocular depth foundation models, benefiting from large-scale synthetic training data, have demonstrated strong generalization. However, they often hal

HapticVLA: Contact-Rich Manipulation via Vision-Language-Action Model without Inference-Time Tactile Sensing

arXiv:2603.15257v2 Announce Type: replace Abstract: Tactile sensing is a crucial capability for Vision-Language-Action (VLA) architectures, as it enables dexterous and safe manipulation in contact-ric

safetyarxiv-cs-ro

HyperODE: Zero-Shot Surrogate for Simulation and Inference of Dynamical Systems

model-releasesarxiv-cs-lg

arXiv:2608.00852v1 Announce Type: new Abstract: Understanding and controlling complex dynamical systems often requires executing thousands of numerical simulations across vast parametric landscapes, w

I built a DwarfStar-inspired Vulkan/Metal inference engine for Qwen3.6-35B-A3B on 16 GB machines

model-releasesr-localllama

Disclosure: I’m the author and maintainer of QuarkStar. I built QuarkStar, a small native inference engine inspired by Antirez’s DwarfStar. QuarkStar currently supports: Qwen3.6-35B-A3B, using the sam

It's the Decoding Format, Not the Perturbation: Auditing Consistency-Based Selection for Vision-Language Test-Time Scaling

arXiv:2608.01207v1 Announce Type: new Abstract: Test-time scaling lifts large language model reasoning by sampling many candidate solutions and selecting among them, yet the same recipe transfers poor

Light Forcing: Accelerating Autoregressive Video Diffusion via Sparse Attention

arXiv:2602.04789v4 Announce Type: replace Abstract: Advanced autoregressive (AR) video generation models have improved visual fidelity and interactivity, but the quadratic complexity of attention rema

MoRAL: Sensor-Grounded BEV Reasoning for Compact VLMs toward Edge-Oriented Autonomous Driving

arXiv:2608.02449v1 Announce Type: new Abstract: Deploying vision-language models (VLMs) for safety-critical spatial reasoning on resource-constrained autonomous driving platforms requires both compact

PredAct-Bench: Benchmarking Tool-Augmented Dialogue under Controlled Tool Noise

arXiv:2608.02372v1 Announce Type: new Abstract: Large Language Models (LLMs) are increasingly deployed in task-oriented dialogue systems that support multi-step decision-making in high-stakes domains