Search: “benchmark”

Subspace Inference Enables Efficient Active Reward Learning from Preferences

arXiv:2609.04066v1 Announce Type: cross Abstract: Reinforcement learning from human feedback (RLHF) has emerged as a powerful yet sample-inefficient approach for learning reward models from human pref

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Model Releases

Synthetic Semantic Supervision for Contrastive Code Representation Learning in Small Transformers: An Empirical Study

DGX agent

arXiv:2609.03702v1 Announce Type: new Abstract: General-purpose code embeddings power tools for code search, classification, and retrieval. Compact transformer encoders for code typically rely on eith

model-releasesarxiv-cs-ai

4 Sept 2026

Local Ai

Temperature Scaling Attack Disrupting Model Confidence in Federated Learning

arXiv:2602.06638v3 Announce Type: replace-cross Abstract: Predictive confidence serves as a foundational control signal in mission-critical systems, directly governing risk-aware logic such as escalat

local-aiarxiv-cs-ai

The impact of phase information for few-shot fine-grained image classification

arXiv:2609.03829v1 Announce Type: cross Abstract: Few-shot fine-grained image classification (FSFGIC) aims to classify similar images with limited labeled examples. This work highlights the critical y

The Impact of Synthetic Data Augmentation on Discourse-Pragmatic Function Classification

model-releasesarxiv-cs-cl

arXiv:2609.03652v1 Announce Type: new Abstract: Synthetic data augmentation has become a common strategy for addressing class imbalance in NLP, but most approaches focus on the quantity and diversity

The Shape of Time: Video-Token Contrast for Temporal Understanding in VideoLMs

arXiv:2609.04110v1 Announce Type: new Abstract: Seeing frames in order does not mean representing time. Modern VideoLMs receive ordered video streams, yet their main supervision acts on generated text

</think> Doesn't Stop Reasoning: Analysis of Spurious CoT Termination

arXiv:2609.03633v1 Announce Type: cross Abstract: Chain-of-thought (CoT) reasoning improves large reasoning models (LRMs) on complex tasks but often produces long, redundant traces. Recent training-fr

TokenMatch: 3D Mesh Correspondence Transformer with Curvature-Guided Tokenisation

arXiv:2609.04202v1 Announce Type: new Abstract: While data-driven 3D shape correspondence estimation has recently seen substantial progress, robust matching under partial observations and strong non-i

Toward Physically Grounded JEPA World Models for Goal-Conditioned Robotic Planning

arXiv:2609.03565v1 Announce Type: cross Abstract: Action-conditioned JEPA world models enable planning toward visually specified goals without reconstructing future pixels, yet latent prediction alone

Towards Multi-modal Multi-turn Safety: From Agentic Interaction to Strategic Alignment

arXiv:2601.04736v2 Announce Type: replace Abstract: Despite remarkable capability in multi-modal understanding, deploying Multi-modal Large Language Models (MLLMs) in open-ended conversational scenari

Towards Numerical TOHTN Planning with SMT-based HTN-SAT Encoding

arXiv:2609.03938v1 Announce Type: new Abstract: While HTN planning has received significant attention in recent years, support for numerical reasoning remains very limited. In this paper, we investiga

Uncertainty Is Not a Safety Net for Clinical VQA, but Can It Anticipate Model Failure?

arXiv:2606.16583v2 Announce Type: replace Abstract: Safe deployment of clinical vision-language models (VLMs) requires reliable uncertainty estimation (UE): a signal indicating when predictions should

Unfold The World: Factorize 4D Properties in Reinforcing Spatial Reasoning

arXiv:2609.03729v1 Announce Type: new Abstract: Despite the remarkable prowess of Vision-Language Models (VLMs) in general multimodal tasks, they remain fundamentally ``flat'' when reasoning about the

safetyarxiv-cs-cv

Agents

Unlocking Lossless Speedups in LLMs via Discrete Diffusion

arXiv:2609.04010v1 Announce Type: new Abstract: Large Language Models (LLMs) owe much of their success to next-token prediction (NTP), but their autoregressive (AR) structure requires slow, sequential

agentsarxiv-cs-lg

User Perceptions vs. Proxy LLM Judges: Privacy and Helpfulness in LLM Responses to Privacy-Sensitive Scenarios

arXiv:2510.20721v4 Announce Type: replace-cross Abstract: Large language models (LLMs) are rapidly being adopted for tasks like drafting emails, summarizing meetings, and answering health questions. I

safetyarxiv-cs-ai

v0.4.0

model-releasesllama-cpp-releases

Overview llama.cpp 0.4.0 adds initial Qwen3.8-Flash-Next and Nemotron-3-Puzzle support, on-demand tensor reading, per-slot server context limits, video input options, and a ggml update to 0.23.0 with

Verify Before You Distill: Prompt-Level Teacher Gating for On-Policy Distillation

arXiv:2609.02998v1 Announce Type: cross Abstract: On-policy distillation (OPD) accelerates post-training by providing dense token-level supervision from a frozen teacher on the student's own rollouts.

What Matters for Aggressive Decoding-Time KV Eviction? Temporal Aggregation and Ranking Preservation

arXiv:2609.03515v1 Announce Type: new Abstract: Decoding-time KV cache compression research focuses heavily on designing better token scoring functions, while the temporal rule that aggregates scores

When Depth Hurts: Reliability-Aware Geometry Distillation for Depth-Free RGB-D Salient Object Detection

arXiv:2609.03378v1 Announce Type: new Abstract: Depth can resolve appearance ambiguity in RGB-D salient object detection (SOD), yet sensor depth is not uniformly reliable. Missing regions, blurred bou

WIDE: Wildcard Inference with Dynamic Expansion for Cross-Modal Generative Retrieval

arXiv:2609.03554v1 Announce Type: cross Abstract: Generative retrieval has demonstrated significant success by unifying representation learning and search into a single sequence-to-sequence generation

Hardware

A Computational Comparison of Fourier Spectral Differentiation and Spatial Automatic Differentiation in Periodic Physics-Informed Neural Networks

arXiv:2609.02110v1 Announce Type: new Abstract: Physics-informed neural networks (PINNs) commonly evaluate the spatial derivatives appearing in partial differential equation residuals using automatic

hardwarearxiv-cs-lg

A Power Law in Logarithm's Clothing: On the Scalability of Graph-Based Vector Search

arXiv:2609.02143v1 Announce Type: cross Abstract: Most vector databases rely on graph-based indexes, notably HNSW and Vamana, for approximate nearest neighbor search. With embedding models widely adop

A Study of Conditional Diffusion Models for Open-Loop Control under Dry Friction and Stiction

model-releasesarxiv-cs-lg

arXiv:2609.01756v1 Announce Type: new Abstract: Diffusion models have recently emerged as expressive generative priors for planning and control. This paper studies Action Diffusion, an action-sequence

A Survey of Transformer-based Language Models with Focus on Efficiency

arXiv:2406.16893v3 Announce Type: replace-cross Abstract: The emergence of Transformer-based Large Language Models (LLMs) has substantially augmented the capabilities of Natural Language Processing (N

Actor as Its Own Critic: Unifying Region Understanding and Localization via CycleGRPO

arXiv:2607.11581v2 Announce Type: replace Abstract: This paper introduces Actor as Its Own Critic, a unified reinforcement learning framework, Cycle Group Relative Policy Optimization (CycleGRPO), tha

AI Contextual Measurement for Recovering Individual and Group-Level Effects: Validation Against Survey Measures and an Occupational Application

arXiv:2609.02821v1 Announce Type: new Abstract: Researchers increasingly use artificial intelligence to construct measures of social, organizational, and occupational characteristics that are absent f

Allocate Before You Embed: Adaptive Visual Input Allocation for Video Embeddings

arXiv:2609.01778v1 Announce Type: new Abstract: Large-scale video retrieval requires embedding models to encode long and diverse videos under tight visual-input and inference budgets. Existing methods

AlphaRAD: Grounded Zero-Shot Classification in Chest Radiology via alpha-Corrected Binary Cross Entropy and Factorized Latent Supervision

arXiv:2609.01757v1 Announce Type: new Abstract: Vision-Language Pretrained Models (VLPMs) offer a scalable path to open-vocabulary chest radiology understanding, yet two aspects remain underexplored:

safetyarxiv-cs-cv

Astra is our most aligned model, with substantial improvements in understanding user intent.

OpenAI announced that its latest model, **Astra**, is their most aligned AI to date, reporting sizable gains in the ability to understand user intent. The tweet was posted on September 3, 2026 at 7:32

model-releasesopenai--x

AutoCompass: Accurate Visual Localization on Public Maps by Learning from Weak Labels

arXiv:2609.02798v1 Announce Type: new Abstract: Neural map matchers estimate an image's 3-DoF pose relative to a 2D map. These models are trained on large-scale datasets of geo-referenced images, whos

Belief-Calibrated Optimization: An Explicit World Model for Agentic Optimization

arXiv:2609.01861v1 Announce Type: new Abstract: The performance of an LLM agent depends on the scaffold around a frozen model. A common way to improve that scaffold is to use a coding agent as an opti

Beyond Appearance: Can Multimodal Large Language Models Exploit Vertical Structure for Remote Sensing Natural Scene Understanding?

arXiv:2605.25784v2 Announce Type: replace Abstract: Multimodal large language models (MLLMs) have advanced rapidly in remote-sensing analysis, yet existing evaluations remain predominantly 2D-centric.

Beyond Textual Chain-of-Thought: A Survey on Action-Grounded Reasoning in Autonomous Driving

arXiv:2609.01659v1 Announce Type: cross Abstract: Chain-of-thought (CoT) reasoning powers generative models by eliciting intermediate steps before producing an answer. In autonomous driving, the answe

Breadth Beats Depth: Improving GCG-Based Jailbreak Optimization with Breadth-Oriented Suffix Search

arXiv:2609.02172v1 Announce Type: new Abstract: Optimization-based jailbreak attacks such as Greedy Coordinate Gradient (GCG) achieve strong effectiveness and transferability by optimizing adversarial

CA-OPD: Confidence-Aware On-Policy Distillation for Structured Visual Prediction

arXiv:2609.02401v1 Announce Type: new Abstract: Autoregressive vision language models unify heterogeneous perception tasks but are highly susceptible to compounding errors. On-policy distillation (OPD

safetyarxiv-cs-cv

Can a strong general-purpose vision model distinguish real from AI-generated images using only a simple decision rule on frozen representati…

model-releasesdavid-ha--x

Can a strong general-purpose vision model distinguish real from AI-generated images using only a simple decision rule on frozen representations? Our latest results, to be presented at #ECCV2026, show

CAPTURE: Disentangling Preference Drift from Memory Poisoning in Personalized LLM Agents

model-releasesarxiv-cs-lg

arXiv:2609.02265v1 Announce Type: new Abstract: Personalized language agents use persistent memory to adapt to users over time, but the same mechanism creates an attack surface. When new information c

Consistency as Regularization for Unsupervised Shadow Removal

arXiv:2609.01806v1 Announce Type: new Abstract: Shadow removal is an important preprocessing step for many vision tasks, yet existing supervised methods require paired shadow and shadow-free images, w

ContextAnyone: Context-Aware Diffusion for Character-Consistent Text-to-Video Generation

arXiv:2512.07328v2 Announce Type: replace-cross Abstract: Text-to-video generation has advanced rapidly, yet preserving a character's holistic appearance from a single reference image remains challeng

CRISP: Cliff-awaRe Input-adaptive Sparse Prefilling with Structural-Mass-Motivated Routing

arXiv:2609.01925v1 Announce Type: cross Abstract: The attention prefilling phase of long-context LLM inference scales quadratically, making self-attention a severe computational bottleneck. Traditiona

researcharxiv-cs-cl

Detecting Object Hallucinations in Large Vision-Language Models via Cross-Modal Attention Drifts and Mask-Based Verification

arXiv:2609.02028v1 Announce Type: new Abstract: Despite recent advances in large vision-language models (LVLMs), object hallucination remains a major barrier to their reliable deployment. Existing det

researcharxiv-cs-cv

DiDrive: A Risk-Aware Hierarchical Diffusion Framework for Safe Offline Reinforcement Learning in Autonomous Driving

model-releasesarxiv-cs-lg

arXiv:2609.01609v1 Announce Type: new Abstract: While diffusion models effectively capture multimodal behavioral priors for autonomous driving, offline reinforcement learning (RL) policies remain susc

DiffIE: Diffusion-based Open Information Extraction

arXiv:2609.02315v1 Announce Type: cross Abstract: A single sentence often expresses multiple valid relational triplets, which makes Open Information Extraction (OpenIE) fundamentally a multi-output ta

Local Ai

DiffuSearch: How Hybrid Trajectory Planning Benefits from Aligned Objectives in Diffusion and Action Space

arXiv:2609.02252v1 Announce Type: cross Abstract: In trajectory planning for autonomous driving, hybrid planning architectures are often realized as a collection of disparate modules, each with its ow

local-aiarxiv-cs-ai

Discriminative World Models for Web Agents

arXiv:2609.02885v1 Announce Type: new Abstract: Recent web agents use world models for test-time action selection by sampling candidate actions, predicting the resulting web states, and ranking them w

Diversifying Long Prompt Image Generation through Structured Prompt Embedding Space Sampling

arXiv:2511.20251v2 Announce Type: replace Abstract: Modern text-to-image models produce impressive visual results from richly specified prompts, yet their behavior under long prompts remains insuffici

DLM-One: Diffusion Language Models for One-Step Sequence Generation

model-releasesarxiv-cs-cl

arXiv:2506.00290v2 Announce Type: replace Abstract: This paper introduces DLM-One, a score-distillation-based framework for one-step sequence generation with continuous diffusion language models (DLMs

Applications

Door-in-the-Face Requests and Refusal Behaviour in Large Language Models

arXiv:2609.02707v1 Announce Type: new Abstract: Does the door-in-the-face technique work on language models? In humans, a large request that is refused makes a smaller follow-up request more likely to

applicationsarxiv-cs-ai