Knowledge catalogue
arXiv:2610.12961v1 Announce Type: cross Abstract: We introduce MVGraph, a benchmark for evaluating large vision-language models (LVLMs) on multi-graph reasoning where the evidence required by a query
arXiv:2610.13390v1 Announce Type: new Abstract: Geospatial machine learning often depends on task-specific feature engineering over large collections of sociodemographic, business, and environmental v
arXiv:2610.13811v1 Announce Type: cross Abstract: Large language models (LLMs) have shown strong results in classical code generation and are increasingly used to write the unit tests that safeguard t
arXiv:2610.12601v1 Announce Type: new Abstract: Prompt optimization is an attractive alternative to reinforcement learning for adapting language-model agents, since it is cheap and produces human-read
arXiv:2610.13823v1 Announce Type: cross Abstract: Semantic query engines ask a large language model (LLM) to judge individual rows, and benchmarks compare these engines on the quality of a few executi
arXiv:2510.13154v3 Announce Type: replace Abstract: What happens inside a language model when alignment training conflicts with a cultural value it encodes? Across 16 MENA countries, 26 models, and 1.
arXiv:2610.13765v1 Announce Type: cross Abstract: Discrete graph diffusion models corrupt graph structure through independent edge updates with a Markovian noise model. Combined with the exchangeabili
A new work suggesting that division of labor among self-improving AI agents can help them avoid reinforcing their own errors. I found this work interesting from a metascience perspective. Science is a
I stumbled across this as I was currently having glm5.3 flash run only around 10tok/s basically unusable. With Maya i am running 30 tok/s now. GLM5.3 feels even with this low quant like a much more en
Alright, I learend a few valuable lessons since I posted about a week ago, and I've made more strides with the DDR4x Ascend 96GB 310P cards I purchased, so here we go, meet the new friendly and less v
Hi, Because sometimes you need both speed AND quality, I made my own Qwen 3.8 27B Swift 1.5 quant. It's an 8 bits AWQ + GPTQ quant. Made with love and precision (read the model card) on my 4 RTX 3090,
Hey everyone! 👋 Feeding multi-gigabyte files (SQL dumps, videos, audio, PDFs, and spreadsheets) directly into LLMs often causes context window overflow, GPU VRAM bottlenecks, massive token bills, and
arXiv:2610.12002v1 Announce Type: cross Abstract: Test-time training (TTT) adapts an LLM's parameters using signals derived from test inputs, and can make striking improvements in pre-specified settin
arXiv:2610.10848v1 Announce Type: new Abstract: Accurate evaluation is central to selecting which LLM to deploy, yet testing a candidate on live traffic exposes real users to an unvetted model. Teams
arXiv:2503.06140v3 Announce Type: replace Abstract: Transfer-based attacks pose a significant threat to real-world applications by directly targeting victim models with adversarial examples generated
arXiv:2509.16525v2 Announce Type: replace-cross Abstract: Machine learning models are increasingly deployed as software components that must evolve as requirements change. When specific training recor
arXiv:2610.12361v1 Announce Type: new Abstract: Large language models increasingly justify legal decisions by naming the statute or precedent behind a verdict, treated as evidence that the decision fo
arXiv:2405.15662v2 Announce Type: replace Abstract: Machine unlearning aims to remove the influence of specified training data or knowledge from a trained model without requiring full retraining. This
arXiv:2610.11447v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly deployed as coding agents that edit files, run builds and tests, inspect execution results, and repair s
arXiv:2604.04611v3 Announce Type: replace Abstract: Federated learning (FL) enables multiple clients to collaboratively train a global model by aggregating local updates without sharing private data.
arXiv:2610.12022v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly deployed in sociotechnical systems where social attribution, the reasoning process attributing external
arXiv:2610.12427v1 Announce Type: cross Abstract: Streaming Video Large Language Models (VLMs) enable continuous video understanding, yet existing benchmarks focus on low-dynamic scenarios. Under boun
arXiv:2610.11005v1 Announce Type: new Abstract: Safety-aligned language models often refuse a harmful request stated directly but answer the same request inside a role-play or narrative wrapper. We me
arXiv:2610.10996v1 Announce Type: new Abstract: Latent reasoning models repeatedly update a latent state using the same recurrent block. As the recurrent depth increases, the sequence of latent states
It's been a long time coming, and I should have done this in the first place. But LumaBrowser is now open source. Since I first released it a year ago, it's had several thousand installs and an absolu
arXiv:2610.11768v1 Announce Type: cross Abstract: Large language model (LLM) agents are beginning to operate industrial energy equipment, and what they get right depends on what they are told about th
arXiv:2610.10655v1 Announce Type: cross Abstract: Large Language Models (LLMs) inevitably internalize substantial amounts of sensitive or private information during pre-training, while LLM unlearning
arXiv:2601.14994v2 Announce Type: replace-cross Abstract: Data contamination can invalidate benchmark evaluation when a model benefits from memorized evaluation content rather than genuine generalizat
arXiv:2610.11460v1 Announce Type: cross Abstract: Prototypes in classification models are almost always vectors, and a vector has no readable form. This paper asks what happens when a prototype is an
arXiv:2610.11183v1 Announce Type: cross Abstract: Following retrieved evidence does not guarantee factual correctness: misleading evidence can induce a model to replace an answer it previously gave co
arXiv:2610.11371v1 Announce Type: new Abstract: Contemporary decoder-only large language models (LLMs) have demonstrated strong capabilities across a wide range of domains. However, existing pretraini
arXiv:2610.10619v1 Announce Type: cross Abstract: Large language model (LLM) agents are rapidly reshaping software engineering, accompanied by an explosion of new code benchmarks. Yet nearly all exist
Figure 1b from our paper Disclosure: I'm one of the authors (Microsoft). The paper, code, dataset are public and ThinkingBox is on Hugging Face OpenEnv as well. Raw evaluation trajectories are not rel
arXiv:2610.12232v1 Announce Type: new Abstract: Test-time adaptation (TTA) methods for time-series forecasting update a deployed model, or a small adapter around it, from incoming ground truth. But th
arXiv:2610.11620v1 Announce Type: new Abstract: Parameter-efficient fine-tuning (PEFT) enables large language models (LLMs) to adapt to specialized tasks, but often at the cost of degrading general ca
arXiv:2610.09505v1 Announce Type: cross Abstract: We develop a bias-aware digital-twin calibration and control framework for multiscale bioprocess models within a biological systems-of-systems (Bio-So
arXiv:2610.10349v1 Announce Type: new Abstract: Instruction-tuned models are deployed into environments where domains are heterogeneous and evolve, yet adding new domains or data typically requires co
model : fix DFlash output head sharing (#30111) llama : fix DFlash output head sharing Assisted-by: Codex dflash : read tied output weights from GGUF metadata Assisted-by: Codex llama : share tied wor
There are way too many forks to keep track of, so I've gotten lost. As far as I can tell, Radiance VLLM is best for models that fit in GPUs while some form of llama.cpp is probably best for MOE RAM-sp
arXiv:2610.08827v1 Announce Type: new Abstract: Automatic Speech Recognition (ASR) systems often underperform for children and non-native speakers, while adapting adult ASR models to child speech can
arXiv:2610.08255v1 Announce Type: cross Abstract: Many organizations adapt large pretrained models to their own tasks by fine-tuning on private data. Several of these parties often hold data for the s
I find it odd watching armchair critics rush to nitpick @MistralAI’s plans or obsess over whether they’re #1 on every leaderboard. Having a team with this level of conviction building frontier models
arXiv:2610.09671v1 Announce Type: new Abstract: Recent advances in reasoning-oriented large language models (LLMs) have motivated increasing evaluation of their ability to perform professional decisio
On-device personal AI requires models and agent software that can operate within fixed hardware limits. Developers also need ways to keep those systems improving after deployment. Model builders are r
arXiv:2610.06647v2 Announce Type: replace Abstract: Reinforcement learning has greatly advanced the capabilities of large language models, but its memory demands remain a barrier to broader adoption.
arXiv:2606.22272v2 Announce Type: replace Abstract: Applying pre-trained language models to new domains through full fine-tuning is computationally expensive and prone to catastrophic forgetting. To a
arXiv:2610.09778v1 Announce Type: cross Abstract: Reproducible benchmarking of Large Language Model (LLM) inference is challenging because repeated measurements can vary with execution and system stat
arXiv:2610.09382v1 Announce Type: cross Abstract: Scribble-based interaction provides a lightweight and intuitive way for users to specify image editing intents in interactive editing tools. However,
arXiv:2610.09074v1 Announce Type: new Abstract: Emerging long-horizon agentic tasks require repeated model calls, worsening the inference cost of already-costly language models. While narrow agentic t
arXiv:2604.01569v2 Announce Type: replace Abstract: Video multimodal large language models achieve strong results on existing benchmarks, but answer accuracy alone does not establish whether they can
arXiv:2610.10197v1 Announce Type: new Abstract: Accurate contact modeling is fundamental to understanding hand-object interaction, yet existing contact representations are typically restricted to obje
arXiv:2610.09313v1 Announce Type: new Abstract: Vision-language models (VLMs) often miss small objects in large images. We ask three questions: what limits them, which of these limits better models ca
Yesterday TechCrunch published a piece about a growing problem for AI agents: websites are starting to block them. Amazon blocking Meta's Muse is the obvious example. At almost exactly the same time,
arXiv:2610.07780v1 Announce Type: new Abstract: Speculative decoding reduces large language model inference latency by drafting multiple tokens before target-model verification, but its effectiveness
arXiv:2601.17645v2 Announce Type: replace-cross Abstract: Internet audio-visual clips convey meaning through time-varying sound and motion, which extend beyond what text alone can represent. To examin
arXiv:2610.07761v1 Announce Type: cross Abstract: In this work, we propose a novel recommendation model, CLARER (Contrastive Learning for Aspect Representation towards Explainable Recommendation) that
arXiv:2610.07928v1 Announce Type: cross Abstract: Dynamic multimodal learning aims to learn robust representations by adaptively modeling information discrepancies across modalities. However, existing
arXiv:2610.08268v1 Announce Type: cross Abstract: Pervasive intelligent applications are increasingly deployed on mobile and Internet of Things (IoT) edge devices. Consequently, Large Language Models
arXiv:2610.07969v1 Announce Type: new Abstract: Scaling robotic foundation models requires diverse training data and reliable evaluation environments. Simulation offers a scalable solution, yet existi
arXiv:2610.08432v1 Announce Type: new Abstract: Improving embodied agents often focuses on optimizing the underlying model through training, while the surrounding agent harness that controls planning,