b11309
Hexagon: optimize ALLREDUCE with support for safe scatter mode (#29757) hex-allreduce: add support for safe scatter mode hex-allreduce: pare down excessive comments hex-allreduce: re-write to remove r
Knowledge catalogue
Hexagon: optimize ALLREDUCE with support for safe scatter mode (#29757) hex-allreduce: add support for safe scatter mode hex-allreduce: pare down excessive comments hex-allreduce: re-write to remove r
cuda: guard the iq4_nl dequantize row kernel against short rows (#29683) dequantize_block_iq4_nl writes QK_K values per block, but a row can be shorter than that (an IQ4_NL row is only guaranteed to b
ggml-opencl : replace alloca() with std::vector (#29765) Signed-off-by: Adrien Gallouët angt@huggingface.co Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/
webgpu: fix SSM_SCAN binding aliasing (#29750) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/51685637 macOS/iOS: macOS Apple Silicon (arm64) macOS Apple S
model : re-enable -sm tensor for qwen4exp (#28569) #27941 disabled -sm tensor for qwen4exp because test-llama-archs asserted on the Meta device once the fixture carried a PLE layer: GGML_ASSERT(ggml_b
metal : use bf16 math for mxfp4 mul-mat (#29770) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/51752863 macOS/iOS: macOS Apple Silicon (arm64) macOS Apple
llama-bench : fix verbosity filter to show GGML_LOG_ERROR (#28229) bench : fix verbosity filter to show GGML_LOG_ERROR (#28107) bench: remove dead variables Website: https://llama.app Attestations: ht
vocab : honor BOS/EOS settings for PLaMo-2 and PLaMo-3 (#29734) vocab : honor BOS/EOS settings for PLaMo-2 and PLaMo-3 The original tokenizer configs for PLaMo-2 and PLaMo-3 have add_bos_token: true a
opencl: mark vec subgroup bcast as supproted for Adreno E17 compiler (#29698) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/51780568 macOS/iOS: macOS Appl
common : add LLM-jp-4.1 Harmony dialect handler (#29681) LLM-jp-4.1 uses the GPT-OSS format, but its tokenizer decodes a space after every special token and parallel tool calls are separated by <|end|
BLAS : Document AOCL-BLAS build and label the device AOCL-BLAS (#29640) BLAS : Document AOCL-BLAS build and label the device AOCL-BLAS AOCL-Blas : Add an AOCL-BLAS Quick Start and drop the fixed versi
hex-workqueue: fix race condition in seqn getting out of sync with idx_read/write (#29785) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/51906249 macOS/iO
HIP: avoid treating CDNA as dgx spark for gqa_ratio 20 in fattn_mma dqk 576 (#29572) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/51919562 macOS/iOS: mac
llama-mmap : avoid a second full-size copy of each tensor with direct-io (#29749) Assisted-by: Claude Co-authored-by: Pranesh Gonegandla pgonegandla@nvidia.com Website: https://llama.app Attestations:
jinja : skip copying loop scope unless a loop filter needs it (#29776) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/51940374 macOS/iOS: macOS Apple Silic
meta: clear inactive AllReduce shards with FILL, not SCALE (#29793) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/51949614 macOS/iOS: macOS Apple Silicon
mtmd: cap max_image to ubatch for non_causal models (#29773) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/51957111 macOS/iOS: macOS Apple Silicon (arm64)
babe, wake up. Hugging Face just dropped a new banger article about RL training coding agents on different harnesses. 1/ Excited to release The ultimate guide to multi-harness RL The same model behave
arXiv:2609.39660v1 Announce Type: new Abstract: Generative models are transforming Bayesian computational imaging, yet the field still lacks physics-aware foundation models. Current practice falls int
arXiv:2609.34211v2 Announce Type: replace Abstract: In financial fraud detection, rich semantic context can provide important evidence for transaction behavior modeling and fraud reasoning. However, p
arXiv:2609.38886v1 Announce Type: new Abstract: Recent advances in robot learning have enabled manipulation policies to perform increasingly diverse tasks and generalize across environments. However,
Hey guys, Looking for an alternative for Claude Cowork: Project Work / Documents Integrations like Notion, Gmail, etc. Tools like Websearch, PDF creation, etc. Came over Eigent (https://github.com/eig
arXiv:2609.39687v1 Announce Type: new Abstract: On-policy self-distillation (OPSD) trains mathematical reasoning models using a privileged teacher that sees a reference solution and supervises student
arXiv:2609.38485v1 Announce Type: new Abstract: Unified multimodal models (UMMs) train image understanding and autoregressive image generation on shared parameters, and the two objectives are known to
arXiv:2609.40117v1 Announce Type: new Abstract: Time-series forecasting models achieve strong benchmark performance but exhibit severe systematic bias in industrial deployments. This train--deploy gap
arXiv:2609.38604v1 Announce Type: cross Abstract: Modern LLM agents increasingly tackle complex tasks through interactive, long-horizon exchanges with users, while existing benchmarks generally assume
arXiv:2609.39751v1 Announce Type: new Abstract: Planning with learned world models combines online trajectory optimization with learned value and policy functions for high-dimensional control. Because
arXiv:2609.39072v1 Announce Type: cross Abstract: Emotion recognition in conversation has been widely studied, but applying Large Language Models (LLMs) to continuous dimensional emotion evaluation in
arXiv:2609.39166v1 Announce Type: new Abstract: Persistent spatial memory enables embodied agents to navigate familiar environments across repeated visits. However, targets may move while unobserved,
arXiv:2609.39646v1 Announce Type: new Abstract: Federated learning faces severe communication bottlenecks when clients upload high-dimensional model updates. Existing methods often compress these upda
arXiv:2609.38660v1 Announce Type: cross Abstract: Long-form subtitle translation requires reasoning over discourse and cultural context spanning episodes or entire series, while maintaining consistent
https://preview.redd.it/5w22g46ihssh1.png?width=2880&format=png&auto=webp&s=cbae24190847b3d80b919ce72d6703c9d1a5a52d The best agent experiences I tried (Dots/ GrokBot) all had an agent that can actual
Every time I needed to explore a dataset quickly, the same wall appeared — cleaning boilerplate, encoding errors, deciding which chart actually makes sense. The mechanical parts aren't interesting, th
Wanted to share a project I've been working on to give local Ollama models persistent document memory without needing heavy vector databases. It's called Hillock. The main problem I had with standard
arXiv:2609.38369v1 Announce Type: new Abstract: We study generalized Blaschke curves as a controlled environment for AI-assisted mathematical rediscovery. For one fixed degree-four Blaschke product, a
arXiv:2609.39394v1 Announce Type: new Abstract: Large language models often solve independent problems in the same conversation. Can computation from earlier problems help them solve new ones? To answ
arXiv:2609.38479v1 Announce Type: cross Abstract: This paper presents an explainable approach to pedestrian routing, in which perceived safety is estimated from street-level imagery through an explici
arXiv:2609.38340v1 Announce Type: new Abstract: When a materials LLM answers a question about crystal structure, does it reason from the structure or copy an answer already printed in its input? Accur
arXiv:2609.38697v1 Announce Type: cross Abstract: We present Cascadia, a system for serving large language models on fleets of commodity Intel AIPCs using their CPU, integrated-GPU, and NPU resources.
arXiv:2609.39441v1 Announce Type: cross Abstract: Online reinforcement learning has been extended to flow matching for diffusion model (DM) image generation. However, this paradigm faces three limitat
arXiv:2609.39533v1 Announce Type: new Abstract: During reinforcement learning with verifiable rewards (RLVR), large language models (LLMs) can exploit loopholes in their environments to obtain high re
arXiv:2609.39629v1 Announce Type: new Abstract: Differential privacy (DP) in machine learning is typically achieved by adding noise to model parameters (private learning) or to model outputs (private
arXiv:2609.38901v1 Announce Type: new Abstract: Representer explanations rank the training landmarks that most influence a self-supervised representation. At scale, this ranking rests on up to four st
arXiv:2609.38781v1 Announce Type: cross Abstract: Multimodal large language models (MLLMs) offer a promising approach for recovering numerical data from scientific charts, but their ability to reconst
arXiv:2609.40190v1 Announce Type: cross Abstract: Sampling several answers and keeping the one a verifier scores highest is one of the simplest ways to buy accuracy at test time. Its effect is reporte
arXiv:2606.16138v2 Announce Type: replace-cross Abstract: Recovering dynamical systems from noisy observations is a recurring challenge across scientific domains, including neuroscience and physics. L
arXiv:2609.39914v1 Announce Type: cross Abstract: Traditional simulations of parametric partial differential equations (PDEs) rely on repetitive computations for each parameter, which makes high-fidel
arXiv:2609.40324v1 Announce Type: new Abstract: We present Cogentic, a multi-agent harness for automated proof discovery on open research problems. While frontier language models can generate strong m
arXiv:2606.19788v2 Announce Type: replace Abstract: We present CombEval, a dynamic benchmark for evaluating combinatorial counting in large language models. CombEval represents each problem as a typed
arXiv:2609.40067v1 Announce Type: cross Abstract: Community Notes is a crowd-sourced approach for adding context to posts on X. Contributors propose and rate notes, forming the inputs to an open-sourc
arXiv:2609.40236v1 Announce Type: new Abstract: Converting free-text radiology reports into structured labels supports cohort building, quality assurance, and monitoring of clinical imaging models, bu
arXiv:2609.33665v2 Announce Type: replace Abstract: Automatically generated environments provide a scalable source of interaction data for training general agents. However, existing approaches mainly
The local models seem powerful enough to be capable of running basic local computer use. This is a computer use agent/harness built via claude code, and powered by qwen3.8 flash next nvfp4. Drawing a
arXiv:2609.38222v1 Announce Type: new Abstract: Retrieval-augmented generation (RAG) can ground large language models in external evidence, but retrieved context does not guarantee that generated clai
arXiv:2609.38891v1 Announce Type: new Abstract: Long-horizon agentic tasks demand strong reasoning and efficient execution across successive interactions with dynamic environments. A common approach d
arXiv:2609.38356v1 Announce Type: new Abstract: Dynamical Systems Reconstruction (DSR) aims to infer models from observed time series that reproduce a system's qualitative long-term behavior. Continua
arXiv:2609.39595v1 Announce Type: new Abstract: Practical Muon maintains momentum and performs a small, fixed number of Newton--Schulz iterations separately for each parameter matrix, often with a Nes
arXiv:2609.39050v1 Announce Type: cross Abstract: As multi-agent systems enter high-stakes domains, the possibility that agents may circumvent safety boundaries is a growing concern. Prior work has ex
arXiv:2609.38208v1 Announce Type: cross Abstract: Smoothed Particle Hydrodynamics (SPH) is well suited to a range of problems, particularly those involving large deformations in fluid dynamics. In rec
arXiv:2609.39584v1 Announce Type: cross Abstract: Edge computing has emerged as a critical computing paradigm in modern distributed systems by migrating data processing closer to end users and Interne