b11403
CUDA: use MMVF for thin f16/bf16 mul_mat at small batch size (#29633) CUDA: use MMVF for thin f16/bf16 mul_mat at small batch size Signed-off-by: ynankani ynankani@nvidia.com adjust kernel selection l
Knowledge catalogue
CUDA: use MMVF for thin f16/bf16 mul_mat at small batch size (#29633) CUDA: use MMVF for thin f16/bf16 mul_mat at small batch size Signed-off-by: ynankani ynankani@nvidia.com adjust kernel selection l
metal : few-row MMA mat-mul (#29869) metal : few-row MMA mat-mul and batched copies for speculative decoding Speculative decoding verifies a few draft tokens per step. Without the tensor API, Metal ra
cuda: tile the lightning indexer over keys and tokens for 4 heads (#29901) cuda: tile the lightning indexer over keys and tokens for 4 heads With too few heads for a wmma tile, a block scores 64 keys
webgpu: add MMVQ support for Q1_0/Q5_0/Q5_1/Q3_K/Q5_K/Q6_K/MXFP4 (#29483) add supports for q1/q5/q3_k/q5_k/q6_k/mxfp4 of mmvq path Add K_QUANTS_HANDLING macro to q1_0 of mmvq path Website: https://lla
vulkan : Fix undeclared identifiers when -DGGML_VULKAN_RUN_TESTS=ON (#29912) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/52752005 macOS/iOS: macOS Apple
ggml-cpu : add Q8_0 IME1 matrix kernel for SpacemiT X60 (#28479) ggml-cpu : add Q8_0 IME1 matrix kernel for SpacemiT X60 On the SpacemiT X60, IME matrix acceleration only covered Q4_0/Q4_1/Q4_K. Q8_0
kv-cache: fix restoring mismatched KV cache rotation by saving exact rotation metadata (#28498) kv-cache: save exact KV rotation metadata, reject restoring mismatched rotation tests : move the state r
llama : fix unexpected graph reallocation in the k-pool models (#29958) llama : fix unexpected graph reallocation in the k-pool models Both k-pool models built a graph shape that depends on state the
vulkan: sparse flash attention for quantized K/V (#29639) vulkan: sparse flash attention for quantized K/V Assisted-by: Claude vulkan: single-scan sparse FA index compaction The compaction ran one wor
vulkan: fix stale prealloc_y reuse across flash attention and soft_max (#29591) Assisted-by: Claude Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/52808438
server: reject partial media truncation (#24076) server: reject partial media truncation server: keep only the keep_first fix Drop the mtmd test helper change, which no longer builds since clip_image_
CUDA: Optimize accumulation in mmq for NVFP4 type (#29857) ggml_cuda: optimize accumulation in mmq_vec_dot_fp4_fp4_mma for better performance remove whitespace fix: correct indentation in mma_block_sc
server: support vision input for Clef (#29969) server: support vision input for Clef move input_attn_causal to private extend old server_batch::embd server_batch::token::pos to multi dim nits fix abor
vulkan: fix Flash Attention shmem write out of bounds (#29988) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/52864580 macOS/iOS: macOS Apple Silicon (arm6
llama.cpp : bump version to 0.6.0 (#29997) llama.cpp : bump version to 0.6.0 scripts : update summary prompt (#0) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attesta
hexagon: matmul and flash-atten scalability updates (#29974) hexagon: head-parallel flash_attn partitioning for row-split multicore In row-split mode each core computes its output row shard of every M
hexagon: add pool op support (#29995) hexagon: add pool_2d support hexagon: add pool_1d support hex-pool: dma changes hex-pool: Optimize HTP pooling boundaries and DMA pipelining hex-pool: code cleanu
arXiv:2610.02730v1 Announce Type: new Abstract: This paper proposes a new functional approximation approach to reduce Bellman error in high-dimensional dynamic programming and Reinforcement Learning p
arXiv:2610.03500v1 Announce Type: new Abstract: Deep tabular generative models are benchmarked on datasets with tens of thousands of rows; clinical datasets have hundreds. We preregistered and ran a s
arXiv:2610.03387v1 Announce Type: new Abstract: Typed decision models return choices or distributions over answer options supplied at request time. Accuracy with complete options does not establish wh
arXiv:2610.03130v1 Announce Type: cross Abstract: Biological literature retrieval systems are often developed and evaluated using broad biomedical corpora and general-purpose search tasks. However, ma
arXiv:2610.03099v1 Announce Type: cross Abstract: E-commerce videos are information-dense and frequently compared by consumers evaluating products and merchants assessing marketing strategies. However
arXiv:2610.03526v1 Announce Type: cross Abstract: Explainers for Graph Neural Networks (GNNs) are commonly evaluated by their plausibility, i.e., how well their explanations recover a predefined groun
arXiv:2511.03756v2 Announce Type: replace-cross Abstract: We present a bifidelity Karhunen--Loeve expansion (KLE) surrogate model for field-valued quantities of interest (QoIs) under uncertain inputs.
arXiv:2610.02800v1 Announce Type: new Abstract: Speculative decoding accelerates autoregressive generation by using a lightweight draft to propose multiple tokens for parallel verification. However, e
arXiv:2610.03193v1 Announce Type: new Abstract: The application of machine learning to dermatology has grown substantially in recent years, moving beyond proof-of-concept studies toward potential appl
arXiv:2610.03162v1 Announce Type: cross Abstract: 3D Gaussian Splatting achieves excellent visual quality with real-time rendering, but at the scale of entire cities it does not fit: a trained model c
arXiv:2603.13856v3 Announce Type: replace-cross Abstract: Building AI systems that can plan, act, and create in the physical world requires more than pattern recognition. Such systems must reason abou
arXiv:2610.03558v1 Announce Type: cross Abstract: Cephalonauts One is a whole-brain 3 Tesla (3T) functional magnetic resonance imaging (fMRI) dataset recorded while subjects listened to audio podcasts
arXiv:2610.02711v1 Announce Type: cross Abstract: Human activity recognition (HAR) from wrist-worn accelerometers is increasingly used for health and behavioral tracking. Yet, most wearable HAR models
arXiv:2602.01124v4 Announce Type: replace Abstract: Dynamic graph representation learning requires capturing both structural relations and temporal evolution, yet existing approaches face a core trade
arXiv:2607.00711v2 Announce Type: cross Abstract: Large Language Models have emerged as programming assistants. However, the efficacy of code generation is constrained by the quality of input requirem
arXiv:2610.02366v1 Announce Type: new Abstract: Finding an optimal behaviour policy within a given environment is a widely studied problem in domains as diverse as games, robotics, energy infrastructu
arXiv:2610.02701v1 Announce Type: new Abstract: Mixture-of-Experts (MoE) models can increase parameter capacity without proportionally increasing active computation, but it is unclear how this trade-o
arXiv:2610.02909v1 Announce Type: new Abstract: When generating programs with language models, constrained decoding can apply program analyses to exclude tokens that violate syntax, scope, or typing r
arXiv:2610.02657v1 Announce Type: new Abstract: Converting a pretrained autoregressive (AR) model to a diffusion language model (dLLM) enables parallel generation without pretraining a new model. Publ
arXiv:2610.01769v1 Announce Type: cross Abstract: Coding agents can generate code that appears correct but implements behavior the user never intended. This mismatch can arise when an agent silently r
arXiv:2610.02780v1 Announce Type: new Abstract: Diffusion models can reach useful sample quality before copying training examples, but fast optimization can compress this generalization window by acce
arXiv:2610.03252v1 Announce Type: new Abstract: Generative models of human-object interaction are bounded by the data that exists: everyday activities involve several objects, but most captured datase
arXiv:2610.02444v1 Announce Type: cross Abstract: Large language models often solve a theorem forward yet fail to disprove a closely related false one: a falsification gap that supervised fine-tuning
arXiv:2610.02527v1 Announce Type: cross Abstract: Learned visual reward models are increasingly used to optimize robot policies, yet a reward model can score an execution that acts on the wrong object
arXiv:2610.02750v1 Announce Type: new Abstract: We address the Person Goal Navigation (PersonNav) problem, enabling robots to search for people under realistic constraints on information accessibility
arXiv:2610.03383v1 Announce Type: new Abstract: Attack Path (AP) modeling is fundamental to cybersecurity analysis, where the Planning Domain Definition Language (PDDL) has been widely adopted to enco
arXiv:2610.03226v1 Announce Type: cross Abstract: GPU kernels generated by large language model (LLM) agents can remain less efficient than expert implementations, but runtime alone does not reveal ho
arXiv:2610.02687v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly deployed in enterprise, scientific, and medical applications, where agents must incorporate domain-specifi
arXiv:2607.19060v2 Announce Type: replace-cross Abstract: Fast prediction of the response of adhesive soft viscoelastic contacts represents a current challenge in soft robotics and for gripping and ma
DeepSeek V4.1 Flash is now available for training on both the Dedicated Training API and Managed Training surfaces! It's a strong base for agentic coding, terminal automation, and tool use. It's also
arXiv:2610.02691v1 Announce Type: new Abstract: Interactive world simulators can provide scalable environments for robot planning, policy training, and evaluation by predicting action consequences whi
arXiv:2602.16543v2 Announce Type: replace Abstract: Safe reinforcement learning (Safe RL) learns robotic controllers that optimize task rewards under safety constraints, yet observation perturbations
arXiv:2610.02351v1 Announce Type: new Abstract: ReAct-based agents typically rely on a single LLM policy to propose actions, interact with the environment, and decide when a task is complete. This cou
arXiv:2610.02631v1 Announce Type: new Abstract: Post-deployment feedback from users can be a cost-effective, scalable, and representative means to monitor and improve generative AI systems and feature
arXiv:2610.03278v1 Announce Type: new Abstract: As dexterous hands proliferate, collecting data and training policies separately for every morphology becomes increasingly impractical. Scalable cross-e
arXiv:2610.02716v1 Announce Type: new Abstract: Objective perturbation adds a random linear term to a regularized empirical risk and releases the exact perturbed minimizer. We study the finite computa
arXiv:2606.17454v3 Announce Type: replace Abstract: AI agent performance is not just a modeling problem, it is fundamentally a systems problem. The advanced capabilities of models are realized through
arXiv:2610.02574v1 Announce Type: new Abstract: High-fidelity solubility prediction is fundamental to pharmaceutical development and environmental partitioning, where accurate modeling must couple mol
arXiv:2610.02866v1 Announce Type: new Abstract: Financial time-series pretraining typically learns from masked observations, contrastive relations, or future outcomes---yet existing objectives struggl
arXiv:2610.03639v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used to support legal practice, education, and research, yet their reliability in national legal systems o
arXiv:2610.03530v1 Announce Type: new Abstract: LiDAR-based quadrotor navigation in cluttered environments remains challenging under external disturbances, particularly when obstacle-aware motion gene
arXiv:2610.03020v1 Announce Type: new Abstract: Long-term agents must remember not only what is true about a user, but also how a particular agent should work with that user as their shared history ev
arXiv:2610.02762v1 Announce Type: new Abstract: Dynamic LLM routers promise to cut inference costs by sending each query to the cheapest model that can answer it correctly. We analyze six commercial r