Search: “model-releases”

Towards Real-World Document Parsing via Realistic Scene Synthesis and Document-Aware Training

arXiv:2603.23885v3 Announce Type: replace Abstract: Document parsing has recently advanced with multimodal large language models (MLLMs) that directly map document images to structured outputs. Tradit

Model Releases

TowerDataset: A Heterogeneous Benchmark for Transmission Corridor Segmentation with a Global-Local Fusion Framework

DGX agent

arXiv:2604.16848v1 Announce Type: new Abstract: Fine-grained semantic segmentation of transmission-corridor point clouds is fundamental for intelligent power-line inspection. However, current progress

model-releasesarxiv-cs-cv

21 Apr 2026

ToxiFrench: Benchmarking and Enhancing Language Models via CoT Fine-Tuning for French Toxicity Detection

arXiv:2508.11281v3 Announce Type: replace Abstract: Detecting toxic content using language models is crucial yet challenging. While substantial progress has been made in English, toxicity detection in

Training for Compositional Sensitivity Reduces Dense Retrieval Generalization

arXiv:2604.16351v1 Announce Type: cross Abstract: Dense retrieval compresses texts into single embeddings ranked by cosine similarity. While efficient for recall, this interface is brittle for identit

TransXion: A High-Fidelity Graph Benchmark for Realistic Anti-Money Laundering

arXiv:2604.17420v1 Announce Type: new Abstract: Money laundering poses severe risks to global financial systems, driving the widespread adoption of machine learning for transaction monitoring. However

Tri-Modal Fusion Transformers for UAV-based Object Detection

arXiv:2604.16630v1 Announce Type: new Abstract: Reliable UAV object detection requires robustness to illumination changes, motion blur, and scene dynamics that suppress RGB cues. Thermal long-wave inf

TriangleMix: Accelerating Prefilling via Decoding-time Contribution Sparsity

arXiv:2507.21526v3 Announce Type: replace Abstract: Large Language Models (LLMs) incur quadratic attention complexity with input length, creating a major time bottleneck in the prefilling stage. Exist

Triples and Knowledge-Infused Embeddings for Clustering and Classification of Scientific Documents

arXiv:2601.08841v2 Announce Type: replace Abstract: The increasing volume and complexity of scientific literature demand robust methods for organizing and understanding research documents. In this stu

TriTS: Time Series Forecasting from a Multimodal Perspective

arXiv:2604.16748v1 Announce Type: new Abstract: Time series forecasting plays a pivotal role in critical sectors such as finance, energy, transportation, and meteorology. However, Long-term Time Serie

TSM-Pose: Topology-Aware Learning with Semantic Mamba for Category-Level Object Pose Estimation

arXiv:2604.16954v1 Announce Type: new Abstract: Category-level object pose estimation is fundamental for embodied intelligence, yet achieving robust generalization to unseen instances remains challeng

TSVer: A Benchmark for Fact Verification Against Time-Series Evidence

arXiv:2511.01101v2 Announce Type: replace Abstract: Reasoning over temporal and numerical data, such as time series, is a crucial aspect of fact-checking. While many systems have recently been develop

UDM-GRPO: Stable and Efficient Group Relative Policy Optimization for Uniform Discrete Diffusion Models

arXiv:2604.18518v1 Announce Type: new Abstract: Uniform Discrete Diffusion Model (UDM) has recently emerged as a promising paradigm for discrete generative modeling; however, its integration with rein

Understanding Tool-Augmented Agents for Lean Formalization: A Factorial Analysis

arXiv:2604.16538v1 Announce Type: cross Abstract: Automatic translation of natural language mathematics into faithful Lean 4 code is hindered by the fundamental dissonance between informal set-theoret

Uni-MMMU: A Massive Multi-discipline Multimodal Unified Benchmark

arXiv:2510.13759v3 Announce Type: replace Abstract: Unified multimodal models aim to jointly enable visual understanding and generation, yet current benchmarks rarely examine their true integration. E

Unified Multimodal Brain Decoding via Cross-Subject Soft-ROI Fusion

arXiv:2512.20249v3 Announce Type: replace-cross Abstract: Multimodal brain decoding aims to reconstruct semantic information that is consistent with visual stimuli from brain activity signals such as

UniMamba: A Unified Spatial-Temporal Modeling Framework with State-Space and Attention Integration

arXiv:2604.16325v1 Announce Type: new Abstract: Multivariate time series forecasting is fundamental to numerous domains such as energy, finance, and environmental monitoring, where complex temporal de

Universally Empowering Zeroth-Order Optimization via Adaptive Layer-wise Sampling

arXiv:2604.18264v1 Announce Type: new Abstract: Zeroth-Order optimization presents a promising memory-efficient paradigm for fine-tuning Large Language Models by relying solely on forward passes. Howe

Unleashing Spatial Reasoning in Multimodal Large Language Models via Textual Representation Guided Reasoning

arXiv:2603.23404v2 Announce Type: replace-cross Abstract: Existing Multimodal Large Language Models (MLLMs) struggle with 3D spatial reasoning, as they fail to construct structured abstractions of the

Unmasking the Illusion of Embodied Reasoning in Vision-Language-Action Models

model-releasesarxiv-cs-ro

arXiv:2604.18000v1 Announce Type: new Abstract: Recent Vision-Language-Action (VLA) models report impressive success rates on standard robotic benchmarks, fueling optimism about general-purpose physic

Unpopular opinion: Grok 4.3 is surprisingly good as a chatbot. Might actually cancel my Gemini subscription.

model-releaseselon-musk--x

This post expresses a positive personal assessment of Grok 4.3's chatbot capabilities, suggesting it performs well enough to potentially replace a Gemini subscription. The statement reflects a user's

Unsupervised Discovery of Intermediate Phase Order in the Frustrated J_1-J_2 Heisenberg Model via Prometheus Framework

arXiv:2602.21468v4 Announce Type: replace-cross Abstract: The spin-1/2 J_1-J_2 Heisenberg model on the square lattice exhibits a debated intermediate phase between Neel antiferromagnetic and stripe or

Unveiling Deepfakes: A Frequency-Aware Triple Branch Network for Deepfake Detection

arXiv:2604.17477v1 Announce Type: new Abstract: Advanced deepfake technologies are blurring the lines between real and fake, presenting both revolutionary opportunities and alarming threats. While it

User-Assistant Bias in LLMs

arXiv:2508.15815v3 Announce Type: replace Abstract: Modern large language models (LLMs) are typically trained and deployed using structured role tags (e.g. system, user, assistant, tool) that explicit

Using large language models for embodied planning introduces systematic safety risks

arXiv:2604.18463v1 Announce Type: cross Abstract: Large language models are increasingly used as planners for robotic systems, yet how safely they plan remains an open question. To evaluate safe plann

VADv2: End-to-End Vectorized Autonomous Driving via Probabilistic Planning

arXiv:2402.13243v2 Announce Type: replace Abstract: Learning a human-like driving policy from large-scale driving demonstrations is promising, but the uncertainty and non-deterministic nature of plann

VCORE: Variance-Controlled Optimization-based Reweighting for Chain-of-Thought Supervision

arXiv:2510.27462v2 Announce Type: replace Abstract: Supervised fine-tuning (SFT) on long chain-of-thought (CoT) trajectories has emerged as a crucial technique for enhancing the reasoning abilities of

Video Panels for Long Video Understanding

arXiv:2509.23724v2 Announce Type: replace Abstract: Recent Video-Language Models (VLMs) achieve promising results on long-video understanding, but their performance still lags behind that achieved on

VIDS: A Verified Imaging Dataset Standard for Medical AI

arXiv:2604.17525v1 Announce Type: cross Abstract: Medical imaging AI development is fundamentally dependent on annotated datasets, yet no existing standard provides machine-enforceable validation acro

VLM-3R: Vision-Language Models Augmented with Instruction-Aligned 3D Reconstruction

arXiv:2505.20279v4 Announce Type: replace-cross Abstract: The rapid advancement of Large Multimodal Models (LMMs) for 2D images and videos has motivated extending these models to understand 3D scenes,

Voronoi-guided Bilateral 2D Gaussian Splatting for Arbitrary-Scale Hyperspectral Image Super-Resolution

arXiv:2604.17727v1 Announce Type: new Abstract: Most existing hyperspectral image super-resolution methods require modifications for different scales, limiting their flexibility in arbitrary-scale rec

WeatherArchive-Bench: Benchmarking Retrieval-Augmented Reasoning for Historical Weather Archives

arXiv:2510.05336v2 Announce Type: replace Abstract: Historical archives on weather events are collections of enduring primary source records that offer rich, untapped narratives of how societies have

🔴 We're going live tomorrow for another Come Build with Pinecone session! Join us to build, break, and learn together. Each session, we pic…

model-releasespinecone--x

🔴 We're going live tomorrow for another Come Build with Pinecone session! Join us to build, break, and learn together. Each session, we pick up a real project — building RAG pipelines, wiring up agent

We're open-sourcing FlashKDA — our high-performance CUTLASS-based implementation of Kimi Delta Attention kernels. Achieves 1.72×–2.22× prefi…

model-releaseskimi-moonshot--x

We're open-sourcing FlashKDA — our high-performance CUTLASS-based implementation of Kimi Delta Attention kernels. Achieves 1.72×–2.22× prefill speedup over the flash-linear-attention baseline on H20,

What makes ChatGPT Images 2.0 a state-of-the-art image generation model? Researchers behind the model explain. A thread: Thinking & Intellig…

What makes ChatGPT Images 2.0 a state-of-the-art image generation model? Researchers behind the model explain. A thread: Thinking & Intelligence in ChatGPT Images 2.0, demonstrated by @ayaanzhaque Med

model-releasesopenai--x

What's Left Unsaid? Detecting and Correcting Misleading Omissions in Multimodal News Previews

arXiv:2601.05563v2 Announce Type: replace Abstract: Even when factually correct, social-media news previews (image-headline pairs) can induce interpretation drift: by selectively omitting crucial cont

When Earth Foundation Models Meet Diffusion: An Application to Land Surface Temperature Super-Resolution

arXiv:2604.16841v1 Announce Type: new Abstract: Land surface temperature (LST) super-resolution is important for environmental monitoring. However, it remains challenging as coarse thermal observation

When Helpers Become Hazards: A Benchmark for Analyzing Multimodal LLM-Powered Safety in Daily Life

arXiv:2601.04043v2 Announce Type: replace Abstract: As Multimodal Large Language Models (MLLMs) become an indispensable assistant in human life, the unsafe content generated by MLLMs poses a danger to

When Pretty Isn't Useful: Investigating Why Modern Text-to-Image Models Fail as Reliable Training Data Generators

arXiv:2602.19946v4 Announce Type: replace Abstract: Recent text-to-image (T2I) diffusion models produce visually stunning images and demonstrate excellent prompt following. But do they perform well as

When Spike Sparsity Does Not Translate to Deployed Cost: VS-WNO on Jetson Orin Nano

arXiv:2604.17040v1 Announce Type: new Abstract: Spiking neural operators are appealing for neuromorphic edge computing because event-driven substrates can, in principle, translate sparse activity into

When Text Hijacks Vision: Benchmarking and Mitigating Text Overlay-Induced Hallucination in Vision Language Models

arXiv:2604.17375v1 Announce Type: new Abstract: Recent advances in Vision-Language Models (VLMs) have substantially enhanced their ability across multimodal video understanding benchmarks spanning tem

When Visuals Aren't the Problem: Evaluating Vision-Language Models on Misleading Data Visualizations

arXiv:2603.22368v2 Announce Type: replace Abstract: Visualizations help communicate data insights, but deceptive data representations can distort their interpretation and propagate misinformation. Whi

When W4A4 Breaks Camouflaged Object Detection: Token-Group Dual-Constraint Activation Quantization

arXiv:2604.16855v1 Announce Type: new Abstract: Camouflaged object detection (COD) segments objects that intentionally blend with the background, so predictions depend on subtle texture and boundary c

Where's the raccoon with the ham radio? (ChatGPT Images 2.0)

model-releasessimon-willison

OpenAI released ChatGPT Images 2.0 today, their latest image generation model. On the livestream Sam Altman said that the leap from gpt-image-1 to gpt-image-2 was equivalent to jumping from GPT-3 to G

Who is the richest club in the championship? Detecting and Rewriting Underspecified Questions Improve QA Performance

arXiv:2602.11938v5 Announce Type: replace Abstract: Large language models (LLMs) perform well on well-posed questions, yet standard question-answering (QA) benchmarks remain far from solved. We argue

Wordle 1,767 4/6 ⬛⬛⬛⬛⬛ 🟨⬛⬛⬛⬛ ⬛🟩🟩⬛⬛ 🟩🟩🟩🟩🟩

model-releasesanthropic--x

This post appears to be a Wordle game result shared by Anthropic on X (formerly Twitter), showing a successful solve on puzzle 1,767 completed in 4 attempts. The colored grid indicators (⬛ for incorre

WorldDB: A Vector Graph-of-Worlds Memory Engine with Ontology-Aware Write-Time Reconciliation

arXiv:2604.18478v1 Announce Type: cross Abstract: Persistent memory is the bottleneck separating stateless chatbots from long-running agentic systems. Retrieval-augmented generation (RAG) over flat ve

XOXO: Stealthy Cross-Origin Context Poisoning Attacks against AI Coding Assistants

arXiv:2503.14281v4 Announce Type: replace-cross Abstract: AI coding assistants are widely used for tasks like code generation. These tools now require large and complex contexts, automatically sourced

XRePIT: A deep learning-computational fluid dynamics hybrid framework implemented in OpenFOAM for fast, robust, and scalable unsteady simulations

arXiv:2510.21804v2 Announce Type: replace Abstract: Autoregressive neural surrogates offer computational acceleration for fluid dynamics but inherently suffer from error accumulation and non-physical