Search: “model-releases”

At last my small pebble of contribution has been merged. Thanks to CISC. Support for the model DFM Mimir 1B gguf here: https://huggingface.co/noctrex/DFM-Mimir additional optimization will be added in

arXiv:2609.17152v1 Announce Type: cross Abstract: Vision Transformers (ViTs) are central to most modern vision models, yet obtaining input attributions that are fine-grained, faithful, and stable rema

arXiv:2609.16660v1 Announce Type: new Abstract: Test-time reinforcement learning adapts a model on its own unlabeled test set using majority-vote pseudo-labels and has shown strong results in mathemat

arXiv:2609.16665v1 Announce Type: cross Abstract: Looped Transformers offer a parameter-efficient route to test-time scaling by reusing shared layers for iterative latent reasoning. However, additiona

arXiv:2609.17435v1 Announce Type: new Abstract: We submit MeTRON-FR, a 125M GPT-2 pretrained on 92.47M words of French, to the BabyLM 2026 Strict track. It scores 85.97 +/- 0.17% on QFrBLiMP (a native

arXiv:2609.16900v1 Announce Type: new Abstract: Platform abuse campaigns conceal redirection instructions with emojis, homophones, character decomposition, and redundant symbols, then route users thro

arXiv:2609.16614v1 Announce Type: cross Abstract: Speech-to-speech dialogue models increasingly support persona control, yet existing spoken role-playing benchmarks remain largely character-centric an

arXiv:2603.02227v2 Announce Type: replace-cross Abstract: Learned gates can approximate sparse attention patterns on frozen transformers, yet provide limited benefit over random gates when trained joi

arXiv:2609.16145v1 Announce Type: new Abstract: We study a practical question: can a small correction module fix errors in a frozen language model's outputs without degrading its base capabilities? We

arXiv:2609.07595v2 Announce Type: replace-cross Abstract: The same underlying computational problem is solved across unrelated fields under different names: recursive Bayesian state estimation appears

arXiv:2609.16233v1 Announce Type: cross Abstract: Vision-language models excel at 2D image understanding but remain limited in 3D spatial reasoning. Progress is hindered by limitations in current benc

arXiv:2609.17523v1 Announce Type: new Abstract: We introduce and release ScienceBuddy, an interactive scientific research workspace that brings continually improving scientific agents into researchers

arXiv:2509.07801v5 Announce Type: replace Abstract: Structured information extraction from scientific literature is crucial for capturing core concepts and emerging trends in specialized fields. While

arXiv:2507.23248v2 Announce Type: replace Abstract: Bengali is spoken by more than 230 million people, yet no standardized instrument evaluates large language models (LLMs) on Bengali across the task

arXiv:2609.15998v1 Announce Type: new Abstract: Every large language model (LLM) has behavioral traits and moral preferences that comprise its character. Whether by design or as an emergent property o

I am planning to buy hardware to run local models. I'm not sure what kind of hardware is best and what will I be able to run. I'm torn between getting a used M1 Max 64gb Macbook for €1500, M3 Max 128g

arXiv:2609.16170v1 Announce Type: new Abstract: Prototype reduction replaces a training set with a smaller representation, and the established methods return a finite set of points. We propose Skeleta

Snap is introducing 'Specs Intelligence,' a new AI assistant that can connect other digital accounts to help you with things like work tasks and keeping track of travel information. It seems similar t

arXiv:2609.17040v1 Announce Type: new Abstract: Wild test-time adaptation (WTTA) updates a source model online under small test batches, concurrent distribution shifts, and time-varying class imbalanc

arXiv:2609.17413v1 Announce Type: new Abstract: This paper investigates easy strategies to boost the performance of existing networks for lidar semantic scene completion (SSC) without requiring comple

arXiv:2609.16340v1 Announce Type: new Abstract: Machine translation systems are periodically upgraded to stronger models, but the available preference signal is human post-edits of an older system's o

arXiv:2609.16055v1 Announce Type: cross Abstract: Test-time compute has emerged as a major approach to improving the capabilities of Large Language Models (LLMs). However, existing test-time reasoning

arXiv:2609.16532v1 Announce Type: new Abstract: Continued pretraining (CPT) with data augmentation such as paraphrasing can store inside a large language model (LLM) the knowledge of a small source co

arXiv:2609.17458v1 Announce Type: new Abstract: Table understanding is a core task in document intelligence, encompassing two key subtasks: table reconstruction and table visual question answering (Ta

arXiv:2609.16754v1 Announce Type: cross Abstract: Fine-tuning an aligned language model on narrow, flawed data can induce harmful behavior far outside the training domain, known as emergent misalignme

arXiv:2609.16797v1 Announce Type: new Abstract: Query-based segmentation methods have shown promising potential for surgical instrument segmentation and recognition, which is essential for scene under

arXiv:2609.16864v1 Announce Type: cross Abstract: Vision-language-action (VLA) models have achieved impressive performance in quasi-static manipulation, but struggle in dynamic manipulation tasks beca

TensorRT Edge‑LLM ran Qwen 3.6‑27B on a single NVIDIA Jetson AGX Thor and completed the MLPerf Edge Agentic benchmark in 24 min 36 s—52.33 tokens/s—which is 6.4× faster than the llama.cpp reference (2

arXiv:2609.16229v1 Announce Type: cross Abstract: Machine unlearning aims to remove specific knowledge from a trained large language model (LLM) without retraining from scratch. Existing methods modif

I wanted to test what Qwen 3.8 27B running locally on a single 5090 could do in Row-Bot. I gave it just one prompt: 'I want you to create an animation showing what the current local model can do. Test

the design of http://evaluatorbench.com grew out of a summer project, http://epistemedia.org (which i still need to write about), and runs on @activegraphai the pattern across all three projects is th

arXiv:2609.15991v1 Announce Type: new Abstract: Standard subword tokenizers either treat every orthographic variation of a word (such as hello, Hello, HELLO, and Hello) as unrelated vocabulary entries

The hardest part of and “org harness” or “enterprise ai” (as opposed to personal ai” is auth, and managing auth in a multiplayer channel (like slack) We’re thinking a lot about how to do this right fo

arXiv:2609.16073v1 Announce Type: cross Abstract: Retrieval-Augmented Generation (RAG) serves as the primary memory architecture for long-horizon autonomous agents. However, treating shared memory as

the independence of AI evaluators is going to matter more... so i made it a 'benchmark': http://evaluatorbench.com (research preview) source-linked directory of third-party evaluators of frontier AI.

arXiv:2609.16745v1 Announce Type: cross Abstract: Action Chunking Transformers (ACT) are widely used to learn robot manipulation from demonstrations. Their conditional variational autoencoder includes

arXiv:2609.16443v1 Announce Type: cross Abstract: State-of-the-art visual locomotion controllers are increasingly capable at handling complex visual environments, making evaluating their real-world pe

arXiv:2609.16247v1 Announce Type: new Abstract: Large language models sometimes behave in ways resembling human emotional responses, and recent work has identified internal representations that may ex

arXiv:2609.16705v1 Announce Type: new Abstract: Physical AI requires more than increasingly large robot datasets: intelligent robots acquire knowledge through continuous interaction with the physical

arXiv:2607.09709v2 Announce Type: replace Abstract: Post-training a code generator against a learned judge can optimize proxy features that raise the score without improving the artifact. We study the

arXiv:2603.21676v2 Announce Type: replace-cross Abstract: Standard Transformers have a fixed computational depth, limiting their ability to generalize to tasks that require variable-depth reasoning. T

ThinkingData Information Technology Pte. Ltd. today announced the launch of the Agentic Engine, an autonomous growth platform for consumer and gaming businesses that uses artificial intelligence to tr

This film cost 29,575 and took 10 days to make. AI ≠ Cheap Our Nexus feature will still cost millions next year. But it'll look like a 200M film. I have a Claude skill you can copy that makes this eas

arXiv:2609.16748v1 Announce Type: new Abstract: Text summarization requires models to condense content while preserving key qualities such as consistency and coherence. Large language models (LLMs) ha

Mistral AI announced on 16 September 2026 that it has partnered with Mozilla to embed privacy, control, and choice features for users interacting with artificial‑intelligence–powered browsing tools. T

arXiv:2609.17317v1 Announce Type: new Abstract: The use of LLMs to complete online surveys impacts the validity of survey-based research, but detecting such usage remains underexplored. We introduce a