Search: “benchmark”

KVMem: Virtualizing Million-Token Agent Workspaces on a Consumer GPU

arXiv:2609.04852v1 Announce Type: new Abstract: Modern LLM agents operate in persistent workspaces whose accumulated history can exceed both GPU KV capacity and the model's native context window. Exis

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Safety

Large Language Models for HVAC Operations in Building Energy Systems: A Critical Review of Methods, Applications, and Deployment Readiness

DGX agent

arXiv:2609.05314v1 Announce Type: new Abstract: Building automation systems generate rich sensor data yet remain insight-poor because heterogeneous point naming, missing metadata, and fragmented docum

safetyarxiv-cs-ai

7 Sept 2026

Latent-Aligned Reasoning for Multimodal Recommendation

arXiv:2609.04645v1 Announce Type: cross Abstract: Multimodal Vision-Language Models (VLMs) have demonstrated remarkable capabilities in cross-modal understanding, yet a fundamental challenge persists

safetyarxiv-cs-cl

Learning-Augmented Algorithms: Guarantees, Construction Mechanisms, and System-Level Implications

arXiv:2609.04787v1 Announce Type: new Abstract: Learning-augmented algorithms use fallible predictions while retaining formal performance guarantees. This survey synthesizes prediction interfaces, err

researcharxiv-cs-lg

Learning Spatial-Spectral Refinement and Calibrating Complementary Observations for Hyperspectral Image Super-Resolution

arXiv:2609.05303v1 Announce Type: new Abstract: Hyperspectral and multispectral image fusion (HMIF) aims to reconstruct a high-resolution hyperspectral image (HR-HSI) by combining the fine spatial det

LentEx: Generalizable Latent Entity Extraction via Synthetic Data and Instruction-Tuned LLMs

model-releasesarxiv-cs-cl

arXiv:2609.04511v1 Announce Type: new Abstract: Latent entity extraction (LEE) tackles the challenge of identifying implicit, contextually inferred entities within free text-an area where traditional

LLM-Guided Program Evolution for Circle Packing: Breaking 10 Packomania Records for $28

arXiv:2609.05093v1 Announce Type: new Abstract: We present Discovery Loop, a lightweight system that uses a large language model (LLM) to iteratively evolve optimization algorithms. Starting from a si

Applications

LUMIN: Lightweight Universal Manufacturing Inspection Network for Anomaly Detection

arXiv:2609.04775v1 Announce Type: new Abstract: Industrial anomaly detection faces two engineering bottlenecks: memory bank construction latency and inference efficiency. Traditional sampling algorith

applicationsarxiv-cs-cv

MABPD: Multi-Agent Bias Probing & Detection via Structured Argument Debate

arXiv:2609.04841v1 Announce Type: cross Abstract: Media bias in news articles operates through subtle linguistic cues---loaded language, selective framing, and strategic omission---that resist single-

MaxKernel: Agentic Kernel Generation for TPUs

arXiv:2609.04523v1 Announce Type: new Abstract: Designing and authoring high-performance custom kernels for accelerators is a complex task that requires deep hardware-level expertise. Large Language M

MINT: A Unified Model for World-Space Camera and Hand Motion Estimation from Scalable Egocentric Pipeline Supervision

arXiv:2609.04958v1 Announce Type: new Abstract: Recovering camera and hand motion in world coordinates from egocentric video is a key capability for activity understanding, robot learning, and augment

researcharxiv-cs-cv

MultiAttenGastro: Multi-Dimensional Attention Augmentation for Gastrointestinal Endoscopy Classification

arXiv:2609.05070v1 Announce Type: new Abstract: Automated gastrointestinal (GI) endoscopy classification requires models that generalize across diverse modalities and class distributions, often far fr

MURAL: Multimodal Uncertainty-aware Recommendation via Adaptive edge Learning

arXiv:2609.04574v1 Announce Type: cross Abstract: Multimodal Graph Neural Networks have become standard for recommendation by augmenting sparse interaction data with content features. Yet current arch

safetyarxiv-cs-lg

NEAT-POCKET: Pocket-Conditioned Autoregressive 3D Molecular Generation with a Neighborhood-Guided Set Transformer

arXiv:2609.05097v1 Announce Type: cross Abstract: AI-driven de novo molecular design offers a promising route to accelerate early-stage drug discovery by generating novel ligands directly within targe

researcharxiv-cs-ai

Nested Inductive Bias Framework for SPD Manifold Learning

arXiv:2609.04466v1 Announce Type: new Abstract: In Geometric Deep Learning, inductive biases serve two primary functions: enforcing manifold constraints and embedding relational priors. Currently, rep

safetyarxiv-cs-lg

NS-ST-GraphRAG: Neuro-Symbolic Spatio-Temporal GraphRAG for Literary Knowledge Processing

model-releasesarxiv-cs-cl

arXiv:2609.05139v1 Announce Type: new Abstract: Long-form literary narratives pose a distinctive information-processing challenge for retrieval-augmented generation: relevant evidence is distributed a

Optimal Rates for Agentic Networked Information Aggregation

arXiv:2609.05318v1 Announce Type: new Abstract: Building on the pioneering paper of Kearns, Roth, and Ryu (SODA'26), we study information aggregation in a networked learning model. The model captures

PAPT++: Risk-Aware Adversarial Tuning and Generation for Single Domain Generalization

arXiv:2609.04837v1 Announce Type: new Abstract: Single domain generalization (SDG) aims to learn a model from one labeled source domain that generalizes to unseen target domains. A common strategy is

safetyarxiv-cs-cv

ParaRNN: Parallel Training for Non-Linear RNNs (sLSTM, RWKV-7, CfC, Titans) via Triton Newton Scans [P]

model-releasesr-machinelearning

I've built and open-sourced ParaRNN, a PyTorch/Triton library that parallelizes the training of non-linear recurrent architectures across time using Newton-Raphson iterations mapped to associative pre

PLUME: Parameter-Efficient Personalization of Large Language Models via Low-Rank User Modulation in Shared Subspaces

arXiv:2609.04715v1 Announce Type: new Abstract: Personalizing large language models (LLMs) is essential for delivering AI assistance that aligns with individual users' styles, intents, and preferences

Post Fusion Bird's Eye View Feature Stabilization for Robust Multimodal 3D Detection

arXiv:2603.05623v2 Announce Type: replace-cross Abstract: Camera-LiDAR fusion is widely used in autonomous driving to enable accurate 3D object detection. However, bird's-eye view (BEV) fusion detecto

ProtLingo: Efficient Protein Language Modeling via Conditional Memory and Expert Routing

arXiv:2609.04793v1 Announce Type: new Abstract: Proteins perform diverse cellular functions, and even single amino-acid substitutions can alter stability, activity, or molecular interactions. Protein

PuTR-CouT: Counting-by-Tracking in Camera-Trap Image Sequences

arXiv:2609.05038v1 Announce Type: new Abstract: Species identification in camera trap images has been widely studied, but key ecological modeling tasks such as species abundance or density estimation

Qwen 3.8 Next Flash is really really REALLY verbose..

model-releasesr-localllama

Long time user of 3.6 27b, switched over to Next Flash since it's a logical step up even from 3.8 27b. It's soooo verbose, i'm talking 13 minutes of thinking time on single turn coding requests at app

Qwen3.8-27B on M1 Max 32GB: MLX 15.8 tok/s vs llama.cpp 9.7 tok/s - but llama.cpp prefill is faster

I’ve been setting up an M1 Max Mac Studio (24-core GPU, 32GB unified memory) as a local LLM server and wanted to compare MLX vs llama.cpp on Qwen3.8-27B. I tried to keep the model footprint and benchm

model-releasesr-ollama

ReCAST: Restoration-aware Cascaded Stage-wise Training for Obfuscated SMS Risk Classification

arXiv:2609.04878v1 Announce Type: cross Abstract: Fraudulent messages sent via Short Message Service (SMS) are increasingly obfuscated to evade cost-conscious classifiers in production systems. In Chi

Reducing Hallucinated Transcripts in Whisper via Hallucination Space Projection

arXiv:2609.04561v1 Announce Type: new Abstract: Whisper is a widely used foundation model for automatic speech recognition (ASR), but its generative decoder can produce fluent hallucinated transcripts

researcharxiv-cs-ai

Risk-Aware Optimal Control with Rulebooks

model-releasesarxiv-cs-ro

arXiv:2609.05199v1 Announce Type: cross Abstract: We consider safety-critical control problems involving multiple requirements with different priorities and uncertainty in their evaluation. We represe

Safety for Whom? Boundary-Aware Self-Distillation for Controlled LLM Safety Refusal

arXiv:2609.04482v1 Announce Type: new Abstract: Safety alignment is usually posed as a topic-level question: is this subject harmful? Deployments ask a narrower one. A civics tutor and a public-sector

safetyarxiv-cs-cl

Same Trajectory, Contradictory Rewards (ROBORMBENCH): Paraphrase Fragility in Vision Language Reward Models

model-releasesarxiv-cs-cl

arXiv:2609.05401v1 Announce Type: cross Abstract: Vision-language models are increasingly used as reward functions for robotic learning, but this role requires paraphrase invariance: the same trajecto

Scalable Context Orchestration for Serving LLMs Over Voice

arXiv:2609.04288v1 Announce Type: cross Abstract: Voice AI applications are gaining popularity as advances in large language models (LLMs) enable more natural and accessible spoken interactions. Servi

safetyarxiv-cs-ai

Scalable Detection of Fossil Palynomorphs in Multifocal Digital Microscopy Images

arXiv:2609.05323v1 Announce Type: new Abstract: Palynomorphs (microscopic, organic-walled fossils such as pollen, spores, and dinoflagellates) are important high-resolution records of past climates an

researcharxiv-cs-cv

Semi-Supervised Hyperspectral Image Classification with Edge-Aware Superpixel Label Propagation and Adaptive Pseudo-Labeling

arXiv:2601.18049v2 Announce Type: replace Abstract: Significant progress has been made in semi-supervised hyperspectral image (HSI) classification regarding feature extraction and classification perfo

SiLR: Structure-Preserving Admission and Process Reward for LLM Tool Agents

arXiv:2609.04629v1 Announce Type: new Abstract: A runtime gate for an LLM tool agent is usually cast as a filter. In a ReAct loop a rejected proposal is followed by another at the same state, so the g

Simulation-free Unbalanced Dynamic Optimal Transport with General Growth Penalty

arXiv:2609.04710v1 Announce Type: cross Abstract: Inferring cellular dynamics from unpaired single-cell snapshots requires modeling both state transitions and population growth or death. Unbalanced dy

researcharxiv-cs-ai

Small Molecule Optimization with Large Language Models

arXiv:2407.18897v2 Announce Type: replace Abstract: Molecular optimization, the process of designing molecules with desirable properties, represents a critical challenge in drug discovery. Recent adva

The microscope is the mask: privileged views and labels from a cryo-ET forward model

arXiv:2609.04325v1 Announce Type: new Abstract: We explore the use of simulated data for training a model for protein annotation in crowded cryo-electron tomography volumes reconstructed from images c

Think-Verify-Revise: Neuro-Symbolic Visual Reasoning with Vision-Language Models and Dynamic Logic Tensor Networks

arXiv:2609.05388v1 Announce Type: new Abstract: Visual reasoning tasks require a system to jointly perceive visual content and apply formal relational constraints---a combination that neither pure neu

This week, three hikers were rescued from a mountain after following AI's advice. On the same day, OpenAI announced that we have reached the era of AGI. The two stories together got me thinking.

On September 1st, three young men from California used Gemini to plan a hiking route on Mount Shasta. The AI told them to bring far less water and food than was actually needed for a hike of that di

model-releasesr-chatgpt

Agents

Trace2Tower: Transition-Aware EigenTrace Induction of Multi-Level Skills for LLM Agents

arXiv:2609.05261v1 Announce Type: new Abstract: Large language model agents increasingly rely on execution traces to master complex interactive tasks. However, current paradigms are bottlenecked by sh

agentsarxiv-cs-ai

Training-Free Speech-Centric Omni Understanding with Frozen VLMs

arXiv:2609.04242v1 Announce Type: cross Abstract: Audio-visual understanding remains challenging because models must jointly interpret spoken content, visual events, and their temporal relationships.

safetyarxiv-cs-cv

Local Ai

TROVE: Adaptive Agent Skill Orchestration via Trace-Grounded Route Validation and Editing

arXiv:2609.05019v1 Announce Type: new Abstract: Agents tend to optimize, select, or constrain execution structures before decisive runtime outcomes are observed. However, such pre-execution commitment

local-aiarxiv-cs-ai

VDN-H3 on M5 Pro 24GB: up to ~2.6× faster H3 generation

model-releasesr-stablediffusion

I recently ported VDN-H3 (Video DeltaNet for MiniMax H3) to Vpipe and tested it on an M5 Pro 24GB. The benefit scales with both sequence length and resolution: at 832×480, VDN-H3 starts to pull ahead

Tutorials

WeakMCN: Multi-task Collaborative Network for Weakly Supervised Referring Expression Comprehension and Segmentation

arXiv:2505.18686v3 Announce Type: replace Abstract: Weakly supervised referring expression comprehension(WREC) and segmentation(WRES) aim to learn object grounding based on a given expression using we

tutorialsarxiv-cs-cv

WEECFP-SuRGE: Wide Embedded Extended Connectivity Fingerprint with Substructure Rotary Graph-distance Encoding

arXiv:2609.04672v1 Announce Type: new Abstract: We introduce WEECFP, a parameter-free 1024-dimensional continuous molecular fingerprint that scatters each Morgan substructure across roughly thirty-two

When Does an Interpretation Count as Established? The Formation, Evaluation, and Responsibility of Interpretation in Generative AI

arXiv:2609.04766v1 Announce Type: cross Abstract: Generative AI research has increasingly evaluated factuality, citation, coverage, and report structure. Yet passing such local checks does not by itse

When Load-Balancing Goes Too Far: Expert Pruning in Over-Dispersed Mixture-of-Experts Models

arXiv:2609.04453v1 Announce Type: cross Abstract: Expert pruning reduces the memory and serving cost of Mixture-of-Experts (MoE) models by removing low-importance experts identified by the router, ass

When Seeing Overrides Knowing: Visual Dominance and Deferral-Based Method for Personalized Safety in VLMs

arXiv:2609.04281v1 Announce Type: cross Abstract: Vision-language models (VLMs) are increasingly deployed in high-stakes settings, where a response that is reasonable in general may still be unsafe fo