Search: “benchmark”

This Reddit thread addresses a common question among Ollama users about whether the tool's perceived slowness is inherent or due to misconfiguration. Ollama's performance is a real concern — multiple

local-air-ollama

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

iTeach: In the Wild Interactive Teaching for Failure-Driven Adaptation of Robot Perception

DGX agent

arXiv:2410.09072v4 Announce Type: replace Abstract: Robotic perception models often fail when deployed in real-world environments due to out-of-distribution conditions such as clutter, occlusion, and

model-releasesarxiv-cs-ro

15 Apr 2026

KG-Hopper: Empowering Compact Open LLMs with Knowledge Graph Reasoning via Reinforcement Learning

arXiv:2603.21440v4 Announce Type: replace-cross Abstract: Large Language Models (LLMs) demonstrate impressive natural language capabilities but often struggle with knowledge-intensive reasoning tasks.

KG-Reasoner: A Reinforced Model for End-to-End Multi-Hop Knowledge Graph Reasoning

arXiv:2604.12487v1 Announce Type: cross Abstract: Large Language Models (LLMs) exhibit strong abilities in natural language understanding and generation, yet they struggle with knowledge-intensive rea

researcharxiv-cs-ai

KnowRL: Boosting LLM Reasoning via Reinforcement Learning with Minimal-Sufficient Knowledge Guidance

arXiv:2604.12627v1 Announce Type: new Abstract: RLVR improves reasoning in large language models, but its effectiveness is often limited by severe reward sparsity on hard problems. Recent hint-based R

Labeled TrustSet Guided: Batch Active Learning with Reinforcement Learning

arXiv:2604.12303v1 Announce Type: new Abstract: Batch active learning (BAL) is a crucial technique for reducing labeling costs and improving data efficiency in training large-scale deep learning model

safetyarxiv-cs-lg

Large Language Models are Powerful Electronic Health Record Encoders

arXiv:2502.17403v5 Announce Type: replace-cross Abstract: Electronic Health Records (EHRs) offer considerable potential for clinical prediction, but their complexity and heterogeneity challenge tradit

Latent Chain-of-Thought World Modeling for End-to-End Driving

model-releasesarxiv-cs-cv

arXiv:2512.10226v2 Announce Type: replace Abstract: Recent Vision-Language-Action (VLA) models for autonomous driving explore inference-time reasoning as a way to improve driving performance and safet

LatentRefusal: Latent-Signal Refusal for Unanswerable Text-to-SQL Queries

arXiv:2601.10398v3 Announce Type: replace Abstract: In LLM-based text-to-SQL systems, unanswerable and underspecified user queries may generate not only incorrect text but also executable programs tha

Learning Chain Of Thoughts Prompts for Predicting Entities, Relations, and even Literals on Knowledge Graphs

arXiv:2604.12651v1 Announce Type: cross Abstract: Knowledge graph embedding (KGE) models perform well on link prediction but struggle with unseen entities, relations, and especially literals, limiting

researcharxiv-cs-ai

Tutorials

LLM-Guided Task- and Affordance-Level Exploration in Reinforcement Learning

arXiv:2509.16615v2 Announce Type: replace Abstract: Reinforcement learning (RL) is a promising approach for robotic manipulation, but it can suffer from low sample efficiency and requires extensive ex

tutorialsarxiv-cs-ro

Long-Horizon Plan Execution in Large Tool Spaces through Entropy-Guided Branching

arXiv:2604.12126v1 Announce Type: new Abstract: Large Language Models (LLMs) have significantly advanced tool-augmented agents, enabling autonomous reasoning via API interactions. However, executing m

LoSA: Locality Aware Sparse Attention for Block-Wise Diffusion Language Models

arXiv:2604.12056v1 Announce Type: new Abstract: Block-wise diffusion language models (DLMs) generate multiple tokens in any order, offering a promising alternative to the autoregressive decoding pipel

researcharxiv-cs-cl

M3D-Stereo: A Multiple-Medium and Multiple-Degradation Dataset for Stereo Image Restoration

model-releasesarxiv-cs-cv

arXiv:2604.12917v1 Announce Type: new Abstract: Image restoration under adverse conditions, such as underwater, haze or fog, and low-light environments, remains a highly challenging problem due to com

Agents

Malice in Agentland: Down the Rabbit Hole of Backdoors in the AI Supply Chain

arXiv:2510.05159v4 Announce Type: replace-cross Abstract: While finetuning AI agents on interaction data -- such as web browsing or tool use -- improves their capabilities, it also introduces critical

agentsarxiv-cs-ai

Mema: Memory-Augmented Adapter for Enhanced Vision-Language Understanding

model-releasesarxiv-cs-cv

arXiv:2603.00655v2 Announce Type: replace Abstract: Multimodal Large Language Models (MLLMs) have achieved remarkable performance by aligning pretrained visual representations with the linguistic know

Memory as Metabolism: A Design for Companion Knowledge Systems

arXiv:2604.12034v1 Announce Type: new Abstract: Retrieval-Augmented Generation remains the dominant pattern for giving LLMs persistent memory, but a visible cluster of personal wiki-style memory archi

MetFuse: Figurative Fusion between Metonymy and Metaphor

arXiv:2604.12919v1 Announce Type: new Abstract: Metonymy and metaphor often co-occur in natural language, yet computational work has studied them largely in isolation. We introduce a framework that tr

researcharxiv-cs-cl

MISID: A Multimodal Multi-turn Dataset for Complex Intent Recognition in Strategic Deception Games

arXiv:2604.12700v1 Announce Type: new Abstract: Understanding human intent in complex multi-turn interactions remains a fundamental challenge in human-computer interaction and behavioral analysis. Whi

Mixed-Density Diffuser: Efficient Planning with Non-Uniform Temporal Resolution

arXiv:2510.23026v5 Announce Type: replace Abstract: Recent studies demonstrate that diffusion planners benefit from sparse-step planning over single-step planning. Training models to skip steps in the

Modality-Native Routing in Agent-to-Agent Networks: A Multimodal A2A Protocol Extension

arXiv:2604.12213v1 Announce Type: new Abstract: Preserving multimodal signals across agent boundaries is necessary for accurate cross-modal reasoning, but it is not sufficient. We show that modality-n

MODIX: A Training-Free Multimodal Information-Driven Positional Index Scaling for Vision-Language Models

arXiv:2604.12537v1 Announce Type: cross Abstract: Vision-Language Models (VLMs) have achieved remarkable progress in multimodal understanding, yet their positional encoding mechanisms remain suboptima

Industry

More people should work on harnesses for open and local models!

Hugging Face co-founder and CEO Clément Delangue advocates for more developers and researchers to focus on building evaluation harnesses and testing frameworks specifically designed for open-source an

industryclem-delangue--x

Agents

M^star: Every Task Deserves Its Own Memory Harness

arXiv:2604.11811v1 Announce Type: cross Abstract: Large language model agents rely on specialized memory systems to accumulate and reuse knowledge during extended interactions. Recent architectures ty

agentsarxiv-cs-ai

Multi-Head Residual-Gated DeepONet for Coherent Nonlinear Wave Dynamics

model-releasesarxiv-cs-lg

arXiv:2604.11972v1 Announce Type: new Abstract: Coherent nonlinear wave dynamics are often strongly shaped by a compact set of physically meaningful descriptors of the initial state. Traditional neura

MultiDocFusion: Hierarchical and Multimodal Chunking Pipeline for Enhanced RAG on Long Industrial Documents

arXiv:2604.12352v1 Announce Type: new Abstract: RAG-based QA has emerged as a powerful method for processing long industrial documents. However, conventional text chunking approaches often neglect com

researcharxiv-cs-ai

MVAdapt: Zero-Shot Multi-Vehicle Adaptation for End-to-End Autonomous Driving

arXiv:2604.11854v1 Announce Type: cross Abstract: End-to-End (E2E) autonomous driving models are usually trained and evaluated with a fixed ego-vehicle, even though their driving policy is implicitly

My bets on open models, mid-2026

A mid-2026 outlook piece from the Interconnects AI newsletter in which the author makes specific predictions about the trajectory of open-weight language models, likely covering expected capability mi

researchinterconnects

Narrative-Driven Paper-to-Slide Generation via ArcDeck

arXiv:2604.11969v1 Announce Type: new Abstract: We introduce ArcDeck, a multi-agent framework that formulates paper-to-slide generation as a structured narrative reconstruction task. Unlike existing m

Navigating the Accuracy-Size Trade-Off with Flexible Model Merging

arXiv:2505.23209v3 Announce Type: replace Abstract: Model merging has emerged as an efficient method to combine multiple single-task fine-tuned models. The merged model can enjoy multi-task capabiliti

researcharxiv-cs-cv

Agents

NaviRAG: Towards Active Knowledge Navigation for Retrieval-Augmented Generation

arXiv:2604.12766v1 Announce Type: new Abstract: Retrieval-augmented generation (RAG) typically relies on a flat retrieval paradigm that maps queries directly to static, isolated text segments. This ap

agentsarxiv-cs-cl

Nemotron 3 Super: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning

arXiv:2604.12374v1 Announce Type: cross Abstract: We describe the pre-training, post-training, and quantization of Nemotron 3 Super, a 120 billion (active 12 billion) parameter hybrid Mamba-Attention

Not All Turns Are Equally Hard: Adaptive Thinking Budgets For Efficient Multi-Turn Reasoning

arXiv:2604.05164v2 Announce Type: replace-cross Abstract: As LLM reasoning performance plateau, improving inference-time compute efficiency is crucial to mitigate overthinking and long thinking traces

NTIRE 2026 The 3rd Restore Any Image Model (RAIM) Challenge: Professional Image Quality Assessment (Track 1)

arXiv:2604.12512v1 Announce Type: cross Abstract: In this paper, we present an overview of the NTIRE 2026 challenge on the 3rd Restore Any Image Model in the Wild, specifically focusing on Track 1: Pr

Applications

Observing the unobserved confounding through its effects: toward randomized trial-like estimates from real-world survival data

arXiv:2604.12137v1 Announce Type: cross Abstract: Background: Randomized controlled trials (RCTs) are costly, time-consuming, and often infeasible, while treatment-effect estimation from observational

applicationsarxiv-cs-ai

Local Ai

Ollama cloud + GLM 5.1 slow and stupid or am I?

This Reddit thread likely discusses user frustrations with performance issues when running GLM-5.1 via Ollama's cloud inference option (`glm-5.1:cloud`), a flagship agentic coding model from Z.AI. Com

local-air-ollama

OmniHands: Towards Robust 4D Hand Mesh Recovery via A Versatile Transformer

arXiv:2405.20330v4 Announce Type: replace-cross Abstract: In this paper, we introduce OmniHands, a universal approach to recovering interactive hand meshes and their relative movement from monocular o

Applications

On Efficient Variants of Segment Anything Model: A Survey

arXiv:2410.04960v5 Announce Type: replace Abstract: The Segment Anything Model (SAM) is a foundational model for image segmentation tasks, known for its strong generalization across diverse applicatio

applicationsarxiv-cs-cv

Operationalising the Right to be Forgotten in LLMs: A Lightweight Sequential Unlearning Framework for Privacy-Aligned Deployment in Politically Sensitive Environments

arXiv:2604.12459v1 Announce Type: new Abstract: Large Language Models (LLMs) are increasingly deployed in politically sensitive environments, where memorisation of personal data or confidential conten

Parametric Interpolation of Dynamic Mode Decomposition for Predicting Nonlinear Systems

model-releasesarxiv-cs-lg

arXiv:2604.12103v1 Announce Type: cross Abstract: We present parameter-interpolated dynamic mode decomposition (piDMD), a parametric reduced-order modeling framework that embeds known parameter-affine

ParetoBandit: Budget-Paced Adaptive Routing for Non-Stationary LLM Serving

model-releasesarxiv-cs-cl

arXiv:2604.00136v2 Announce Type: replace-cross Abstract: Multi-model LLM serving operates in a non-stationary, noisy environment: providers revise pricing, model quality can shift or regress without

Perception-Aware Policy Optimization for Multimodal Reasoning

arXiv:2507.06448v5 Announce Type: replace Abstract: Reinforcement Learning with Verifiable Rewards (RLVR) has proven to be a highly effective strategy for endowing Large Language Models (LLMs) with ro

safetyarxiv-cs-cl

PILOT: Planning via Internalized Latent Optimization Trajectories for Large Language Models

arXiv:2601.19917v2 Announce Type: replace Abstract: Strategic planning is critical for multi-step reasoning, yet compact Large Language Models (LLMs) often lack the capacity to formulate global strate

researcharxiv-cs-cl

Policy-Invisible Violations in LLM-Based Agents

arXiv:2604.12177v1 Announce Type: new Abstract: LLM-based agents can execute actions that are syntactically valid, user-sanctioned, and semantically appropriate, yet still violate organizational polic

Polynomial Expansion Rank Adaptation: Enhancing Low-Rank Fine-Tuning with High-Order Interactions

arXiv:2604.11841v1 Announce Type: cross Abstract: Low-rank adaptation (LoRA) is a widely used strategy for efficient fine-tuning of large language models (LLMs), but its strictly linear structure fund

PR-MaGIC: Prompt Refinement Via Mask Decoder Gradient Flow For In-Context Segmentation

arXiv:2604.12113v1 Announce Type: cross Abstract: Visual Foundation Models (VFMs) such as the Segment Anything Model (SAM) have significantly advanced broad use of image segmentation. However, SAM and

PromptEcho: Annotation-Free Reward from Vision-Language Models for Text-to-Image Reinforcement Learning

arXiv:2604.12652v1 Announce Type: cross Abstract: Reinforcement learning (RL) can improve the prompt following capability of text-to-image (T2I) models, yet obtaining high-quality reward signals remai

Public Profile Matters: A Scalable Integrated Approach to Recommend Citations in the Wild

arXiv:2603.17361v2 Announce Type: replace-cross Abstract: Proper citation of relevant literature is essential for contextualising and validating scientific contributions. While current citation recomm