We present Adjacent Possible Exploration (APE), a selective fine-tuning method for adapting large language models that systematically explores parameter modifications while maintaining model stability. Inspired by evolutionary optimization principles, APE evaluates multiple candidate parameter updates through fine-tuning on small data subsets and accepts only those exceeding a performance threshold. Unlike standard fine-tuning that follows single gradient directions, APE implements a filtered selection process that prevents destabilizing parameter changes while enabling systematic improvement. Our method achieves 33.9\% BLEU improvement and 36.2\% perplexity reduction on news summarization tasks while using minimal computational resources. The approach provides a practical framework for controlled model adaptation that balances performance gains with representational stability.
Inference is now the dominant AI workload, yet existing systems force trade-offs between latency, throughput, and cost. Arctic Inference, an open-source vLLM plugin from Snowflake AI Research, introduces Shift Parallelism, a dynamic parallelism strategy that adapts to real-world traffic while integrating speculative decoding, SwiftKV compute reduction, and optimized embedding inference. It achieves up to 3.4 times faster request completion, 1.75 times faster generation, and 1.6M tokens/sec per GPU for embeddings, outperforming both latency- and throughput-optimized deployments. Already powering Snowflake Cortex AI, Arctic Inference delivers state-of-the-art, cost-effective inference for enterprise AI and is now available to the community.
Transformer models are computationally costly on long sequences since regular attention has quadratic $O(n^2)$ time complexity. We introduce Wavelet-Enhanced Random Spectral Attention (WERSA), a novel mechanism of linear $O(n)$ time complexity that is pivotal to enable successful long-sequence processing without the performance trade-off. WERSA merges content-adaptive random spectral features together with multi-resolution Haar wavelets and learnable parameters to selectively attend to informative scales of data while preserving linear efficiency. Large-scale comparisons \textbf{on single GPU} and across various benchmarks (vision, NLP, hierarchical reasoning) and various attention mechanisms (like Multiheaded Attention, Flash-Attention-2, FNet, Linformer, Performer, Waveformer), reveal uniform advantages of WERSA. It achieves best accuracy in all tests. On ArXiv classification, WERSA improves accuracy over vanilla attention by 1.2\% (86.2\% vs 85.0\%) while cutting training time by 81\% (296s vs 1554s) and FLOPS by 73.4\% (26.2G vs 98.4G). Significantly, WERSA excels where vanilla and FlashAttention-2 fail: on ArXiv-128k's extremely lengthy sequences, it achieves best accuracy (79.1\%) and AUC (0.979) among viable methods, operating on data that gives Out-Of-Memory errors to quadratic methods while being \textbf{twice as fast} as Waveformer, its next-best competitor. By significantly reducing computational loads without compromising accuracy, WERSA makes possible more practical, more affordable, long-context models, in particular on low-resource hardware, for more sustainable and more scalable AI development.
Graph-based Point Cloud Networks (PCNs) are powerful tools for processing sparse sensor data with irregular geometries, as found in high-energy physics detectors. However, deploying models in such environments remains challenging due to stringent real-time requirements for both latency, and throughput. In this work, we present a deeply pipelined dataflow architecture for executing graph-based PCNs on FPGAs. Our method supports efficient processing of dynamic, sparse point clouds while meeting hard real-time constraints. We introduce specialized processing elements for core graph operations, such as GraVNet convolution and condensation point clustering, and demonstrate our design on the AMD Versal VCK190. Compared to a GPU baseline, our FPGA implementation achieves up to 5.25x speedup in throughput while maintaining latencies below 10 {\mu}s, satisfying the demands of real-time trigger systems in particle physics experiments. An open-source reference implementation is provided.
Score-function policy gradients have delivered strong results in game-playing, robotics and language-model fine-tuning. Yet its high-variance often undermines training stability. On the other hand, pathwise policy gradients alleviate the training variance, but are reliable only when driven by an accurate action-conditioned value function which is notoriously hard to train without relying on past off-policy data. In this paper, we discuss how to construct a value-gradient driven, on-policy algorithm that allow training Q-value models purely from on-policy data, unlocking the possibility of using pathwise policy updates in the context of on-policy learning. We show how to balance stochastic policies for exploration with constrained policy updates for stable training, and evaluate important architectural components that facilitate accurate value function learning. Building on these insights, we propose Relative Entropy Pathwise Policy Optimization (REPPO), an efficient on-policy algorithm that combines the sample-efficiency of pathwise policy gradients with the simplicity and minimal memory footprint of standard on-policy learning. We demonstrate that REPPO provides strong empirical performance at decreased sample requirements, wall-clock time, memory footprint as well as high hyperparameter robustness in a set of experiments on two standard GPU-parallelized benchmarks.
Pixel-level annotation is expensive and time-consuming. Semi-supervised segmentation methods address this challenge by learning models on few labeled images alongside a large corpus of unlabeled images. Although foundation models could further account for label scarcity, effective mechanisms for their exploitation remain underexplored. We address this by devising a novel semi-supervised panoptic approach fueled by two dedicated foundation models. We enhance recognition by complementing unsupervised mask-transformer consistency with zero-shot classification of CLIP features. We enhance localization by class-agnostic decoder warm-up with respect to SAM pseudo-labels. The resulting decoupled enhancement of recognition and localization (DEARLi) particularly excels in the most challenging semi-supervised scenarios with large taxonomies and limited labeled data. Moreover, DEARLi outperforms the state of the art in semi-supervised semantic segmentation by a large margin while requiring 8x less GPU memory, in spite of being trained only for the panoptic objective. We observe 29.9 PQ and 38.9 mIoU on ADE20K with only 158 labeled images. The source code is available at https://github.com/helen1c/DEARLi.
In medical image segmentation, convolutional neural networks (CNNs) and transformers are dominant. For CNNs, given the local receptive fields of convolutional layers, long-range spatial correlations are captured through consecutive convolutions and pooling. However, as the computational cost and memory footprint can be prohibitively large, 3D models can only afford fewer layers than 2D models with reduced receptive fields and abstract levels. For transformers, although long-range correlations can be captured by multi-head attention, its quadratic complexity with respect to input size is computationally demanding. Therefore, either model may require input size reduction to allow more filters and layers for better segmentation. Nevertheless, given their discrete nature, models trained with patch-wise training or image downsampling may produce suboptimal results when applied on higher resolutions. To address this issue, here we propose the resolution-robust HNOSeg-XS architecture. We model image segmentation by learnable partial differential equations through the Fourier neural operator which has the zero-shot super-resolution property. By replacing the Fourier transform by the Hartley transform and reformulating the problem in the frequency domain, we created the HNOSeg-XS model, which is resolution robust, fast, memory efficient, and extremely parameter efficient. When tested on the BraTS'23, KiTS'23, and MVSeg'23 datasets with a Tesla V100 GPU, HNOSeg-XS showed its superior resolution robustness with fewer than 34.7k model parameters. It also achieved the overall best inference time (< 0.24 s) and memory efficiency (< 1.8 GiB) compared to the tested CNN and transformer models.
We investigate reinforcement learning (RL) for privileged planning in autonomous driving. State-of-the-art approaches for this task are rule-based, but these methods do not scale to the long tail. RL, on the other hand, is scalable and does not suffer from compounding errors like imitation learning. Contemporary RL approaches for driving use complex shaped rewards that sum multiple individual rewards, \eg~progress, position, or orientation rewards. We show that PPO fails to optimize a popular version of these rewards when the mini-batch size is increased, which limits the scalability of these approaches. Instead, we propose a new reward design based primarily on optimizing a single intuitive reward term: route completion. Infractions are penalized by terminating the episode or multiplicatively reducing route completion. We find that PPO scales well with higher mini-batch sizes when trained with our simple reward, even improving performance. Training with large mini-batch sizes enables efficient scaling via distributed data parallelism. We scale PPO to 300M samples in CARLA and 500M samples in nuPlan with a single 8-GPU node. The resulting model achieves 64 DS on the CARLA longest6 v2 benchmark, outperforming other RL methods with more complex rewards by a large margin. Requiring only minimal adaptations from its use in CARLA, the same method is the best learning-based approach on nuPlan. It scores 91.3 in non-reactive and 90.6 in reactive traffic on the Val14 benchmark while being an order of magnitude faster than prior work.
We present any4, a learned 4-bit weight quantization solution for large language models (LLMs) providing arbitrary numeric representations without requiring pre-processing of weights or activations. any4 yields higher accuracy compared to other related 4-bit numeric representation types: int4, fp4 and nf4, as evaluated on a range of model sizes, generations and families (Llama 2, Llama 3, Mistral and Mixtral). While any4 does not require preprocessing of weights or activations, it is also competitive with orthogonal techniques that require such preprocessing (e.g., AWQ and GPTQ). We also experiment with any3 and any2 and show competitiveness at lower bits. Additionally, we show that we can calibrate using a single curated diverse sample rather than hundreds of samples from a dataset as done in most quantization approaches. We also open source tinygemm, a latency optimized GPU matrix multiplication library for LLMs, that implements any4 using a GPU-efficient lookup table strategy along with other common quantization methods. We open source our code at https://github.com/facebookresearch/any4 .
Kernel development in deep learning requires optimizing computational units across hardware while balancing memory management, parallelism, and hardware-specific optimizations through extensive empirical tuning. Although domain-specific languages like Triton simplify GPU programming by abstracting low-level details, developers must still manually tune critical parameters such as tile sizes and memory access patterns through iterative experimentation, creating substantial barriers to optimal performance and wider adoption. In this work, we introduce AutoTriton, the first model dedicated to Triton programming powered by reinforcement learning (RL). AutoTriton performs supervised fine-tuning (SFT) to be equipped with essential Triton programming expertise using a high-quality data gathering pipeline, and conducts RL with Group Relative Policy Optimization (GRPO) algorithm, combining a rule-based reward and an execution-based reward to further improve Triton programming ability, sequentially. Experiments across five evaluation channels of TritonBench and KernelBench illustrate that our 8B model AutoTriton achieves performance comparable to mainstream large models, including Claude-4-Sonnet and DeepSeek-R1-0528. Further experimental analysis demonstrates the crucial role of each module within AutoTriton, including the SFT stage, the RL stage, and the reward design strategy. These findings underscore the promise of RL for automatically generating high-performance kernels, and since high-performance kernels are core components of AI systems, this breakthrough establishes an important foundation for building more efficient AI systems. The model and code will be available at https://github.com/AI9Stars/AutoTriton.
Large Language Models (LLMs) have achieved remarkable performance on complex mathematical benchmarks, yet often struggle with simple arithmetic tasks and exhibit a tendency toward over-explaining or "overthinking" answers. To systematically assess this phenomenon, we introduce LLMThinkBench, a modular benchmarking framework that enables researchers to evaluate basic math reasoning and overthinking in LLMs. The framework provides 14 configurable math tasks with randomized test data generation and robust parsing strategies. Researchers can quantify overthinking using our Overthinking Score metric, which captures accuracy-verbosity tradeoffs through harmonic mean formulation. The tool offers flexible evaluation with a scalable vLLM/Transformers backend, multi-GPU support, and full configurability. Users can extend the tool with custom tasks, reproduce experiments with seeding, and generate detailed efficiency reports. Distributed as a pip-installable package with CLI and API access, LLMThinkBench provides researchers and practitioners an accessible, cost-effective alternative to expensive LLM-as-a-judge methods for diagnosing basic reasoning capabilities and efficiency analysis. Package can be installed as: pip install llmthinkbench
We present \texttt{MathOptAI.jl}, an open-source Julia library for embedding trained machine learning predictors into a JuMP model. \texttt{MathOptAI.jl} can embed a wide variety of neural networks, decision trees, and Gaussian Processes into a larger mathematical optimization model. In addition to interfacing a range of Julia-based machine learning libraries such as \texttt{Lux.jl} and \texttt{Flux.jl}, \texttt{MathOptAI.jl} uses Julia's Python interface to provide support for PyTorch models. When the PyTorch support is combined with \texttt{MathOptAI.jl}'s gray-box formulation, the function, Jacobian, and Hessian evaluations associated with the PyTorch model are offloaded to the GPU in Python, while the rest of the nonlinear oracles are evaluated on the CPU in Julia. \MathOptAI is available at https://github.com/lanl-ansi/MathOptAI.jl under a BSD-3 license.
Enhancing the network architecture of the YOLO framework has been crucial for a long time, but has focused on CNN-based improvements despite the proven superiority of attention mechanisms in modeling capabilities. This is because attention-based models cannot match the speed of CNN-based models. This paper proposes an attention-centric YOLO framework, namely YOLOv12, that matches the speed of previous CNN-based ones while harnessing the performance benefits of attention mechanisms. YOLOv12 surpasses all popular real-time object detectors in accuracy with competitive speed. For example, YOLOv12-N achieves 40.6% mAP with an inference latency of 1.64 ms on a T4 GPU, outperforming advanced YOLOv10-N / YOLOv11-N by 2.1%/1.2% mAP with a comparable speed. This advantage extends to other model scales. YOLOv12 also surpasses end-to-end real-time detectors that improve DETR, such as RT-DETR / RT-DETRv2: YOLOv12-S beats RT-DETR-R18 / RT-DETRv2-R18 while running 42% faster, using only 36% of the computation and 45% of the parameters. More comparisons are shown in Figure 1.
Residual connection has been extensively studied and widely applied at the model architecture level. However, its potential in the more challenging data-centric approaches remains unexplored. In this work, we introduce the concept of Data Residual Matching for the first time, leveraging data-level skip connections to facilitate data generation and mitigate data information vanishing. This approach maintains a balance between newly acquired knowledge through pixel space optimization and existing core local information identification within raw data modalities, specifically for the dataset distillation task. Furthermore, by incorporating optimization-level refinements, our method significantly improves computational efficiency, achieving superior performance while reducing training time and peak GPU memory usage by 50%. Consequently, the proposed method Fast and Accurate Data Residual Matching for Dataset Distillation (FADRM) establishes a new state-of-the-art, demonstrating substantial improvements over existing methods across multiple dataset benchmarks in both efficiency and effectiveness. For instance, with ResNet-18 as the student model and a 0.8% compression ratio on ImageNet-1K, the method achieves 47.7% test accuracy in single-model dataset distillation and 50.0% in multi-model dataset distillation, surpassing RDED by +5.7% and outperforming state-of-the-art multi-model approaches, EDC and CV-DD, by +1.4% and +4.0%. Code is available at: https://github.com/Jiacheng8/FADRM.
Parametric human body models play a crucial role in computer graphics and vision, enabling applications ranging from human motion analysis to understanding human-environment interactions. Traditionally, these models use surface meshes, which pose challenges in efficiently handling interactions with other geometric entities, such as objects and scenes, typically represented as meshes or point clouds. To address this limitation, recent research has explored volumetric neural implicit body models. However, existing works are either insufficiently robust for complex human articulations or impose high computational and memory costs, limiting their widespread use. To this end, we introduce VolumetricSMPL, a neural volumetric body model that leverages Neural Blend Weights (NBW) to generate compact, yet efficient MLP decoders. Unlike prior approaches that rely on large MLPs, NBW dynamically blends a small set of learned weight matrices using predicted shape- and pose-dependent coefficients, significantly improving computational efficiency while preserving expressiveness. VolumetricSMPL outperforms prior volumetric occupancy model COAP with 10x faster inference, 6x lower GPU memory usage, enhanced accuracy, and a Signed Distance Function (SDF) for efficient and differentiable contact modeling. We demonstrate VolumetricSMPL's strengths across four challenging tasks: (1) reconstructing human-object interactions from in-the-wild images, (2) recovering human meshes in 3D scenes from egocentric views, (3) scene-constrained motion synthesis, and (4) resolving self-intersections. Our results highlight its broad applicability and significant performance and efficiency gains.
Recent advances in optical flow estimation have prioritized accuracy at the cost of growing GPU memory consumption, particularly for high-resolution (FullHD) inputs. We introduce MEMFOF, a memory-efficient multi-frame optical flow method that identifies a favorable trade-off between multi-frame estimation and GPU memory usage. Notably, MEMFOF requires only 2.09 GB of GPU memory at runtime for 1080p inputs, and 28.5 GB during training, which uniquely positions our method to be trained at native 1080p without the need for cropping or downsampling. We systematically revisit design choices from RAFT-like architectures, integrating reduced correlation volumes and high-resolution training protocols alongside multi-frame estimation, to achieve state-of-the-art performance across multiple benchmarks while substantially reducing memory overhead. Our method outperforms more resource-intensive alternatives in both accuracy and runtime efficiency, validating its robustness for flow estimation at high resolutions. At the time of submission, our method ranks first on the Spring benchmark with a 1-pixel (1px) outlier rate of 3.289, leads Sintel (clean) with an endpoint error (EPE) of 0.963, and achieves the best Fl-all error on KITTI-2015 at 2.94%. The code is available at https://github.com/msu-video-group/memfof.
We introduce DIP, a novel unsupervised post-training method designed to enhance dense image representations in large-scale pretrained vision encoders for in-context scene understanding. Unlike prior approaches that rely on complex self-distillation architectures, our method trains the vision encoder using pseudo-tasks that explicitly simulate downstream in-context scenarios, inspired by meta-learning principles. To enable post-training on unlabeled data, we propose an automatic mechanism for generating in-context tasks that combines a pretrained diffusion model and the vision encoder itself. DIP is simple, unsupervised, and computationally efficient, requiring less than 9 hours on a single A100 GPU. By learning dense representations through pseudo in-context tasks, it achieves strong performance across a wide variety of downstream real-world in-context scene understanding tasks. It outperforms both the initial vision encoder and prior methods, offering a practical and effective solution for improving dense representations. Code available here: https://github.com/sirkosophia/DIP
Recent advances in multimodal generative models have unlocked photorealistic, instruction-aligned image generation, yet leading systems like GPT-4o-Image remain proprietary and inaccessible. To democratize these capabilities, we present ShareGPT-4o-Image, the first dataset comprising 45K text-to-image and 46K text-and-image-to-image data, all synthesized using GPT-4o's image generation capabilities for distilling its advanced image generation abilities. Leveraging this dataset, we develop Janus-4o, a multimodal large language model capable of both text-to-image and text-and-image-to-image generation. Janus-4o not only significantly improves text-to-image generation over its predecessor, Janus-Pro, but also newly supports text-and-image-to-image generation. Notably, it achieves impressive performance in text-and-image-to-image generation from scratch, using only 91K synthetic samples and 6 hours of training on an 8 A800-GPU machine. We hope the release of ShareGPT-4o-Image and Janus-4o will foster open research in photorealistic, instruction-aligned image generation.
Self-supervised learning (SSL) models such as WavLM have brought substantial improvements to speaker diarization by providing rich contextual representations. However, the high computational and memory costs of these models hinder their deployment in real-time and resource-constrained scenarios. In this work, we present a comprehensive study on compressing SSL-based diarization models through structured pruning guided by knowledge distillation. Building upon our previous work, we extend the analysis to include pruning objectives based on multiply-accumulate operations (MACs), investigate module-wise and progressive pruning strategies, and examine the impact of training data quantity. Experimental results show that our method reduces model size by up to 80% without degrading performance, achieving up to 4x faster inference on a single GPU. We further perform large-scale evaluations on a diverse compound dataset comprising eight public diarization corpora, where our best pruned model achieves state-of-the-art performance across most conditions. Additionally, we show strong generalization to the CHiME-6 dataset, attaining performance comparable to the third-place system in the CHiME-7 challenge without any domain adaptation. All models and code are publicly released to support reproducibility and future research.
The recent shift in Generative AI (GenAI) applications from cloud-only environments to end-user devices introduces new challenges in resource management, system efficiency, and user experience. This paper presents ConsumerBench, a comprehensive benchmarking framework designed to evaluate the system efficiency and response time of GenAI models running on end-user devices. Unlike existing benchmarks that assume exclusive model access on dedicated GPUs, ConsumerBench simulates realistic multi-application scenarios executing concurrently on constrained hardware. Furthermore, ConsumerBench supports customizable workflows that simulate complex tasks requiring coordination among multiple applications. ConsumerBench captures both application-level metrics, including latency and Service Level Objective (SLO) attainment, and system-level metrics like CPU/GPU utilization and memory bandwidth. Through extensive experiments, ConsumerBench reveals inefficiencies in resource sharing, unfair scheduling under greedy allocation, and performance pitfalls of static model server configurations. The paper also provides practical insights for model developers and system designers, highlighting the benefits of custom kernels tailored to consumer-grade GPU architectures and the value of implementing SLO-aware scheduling strategies.