arXiv Machine Learning
Jul 23
How Fast Can Reward Models Score? A Systems Study of C++ and PyTorch Inference Runtimes for RLHF
arXiv:2607. 19712v1 Announce Type: new Abstract: In RLHF pipelines, reward scoring blocks policy updates.
How Fast Can Reward Models Score? A Systems Study of C++ and PyTorch Inference Runtimes for RLHF
In RLHF pipelines, reward scoring blocks policy updates. Slow scoring bottlenecks the entire loop, since no update runs until every rollout gets a score.
Related stories
arXiv AI
Jul 21
KernelBench-Verified: Do LLM-Generated Kernels Actually Beat PyTorch?
arXiv:2607. 16241v1 Announce Type: cross Abstract: Recent large language models (LLMs) can generate custom CUDA kernels that appear to outperform PyTorch on benchmarks such as KernelBench.
arXiv AI
Sep 21
How Much of a Real Workload Can LLM-Generated GPU Kernels Actually Reach?
arXiv:2609.21058v1 Announce Type: cross Abstract: Language models can now write GPU kernels that outperform PyTorch. We evaluate five model configurations on KernelBench level 1 and find that a front...
arXiv AI
Jun 30
Metal-Sci: A Scientific Compute Benchmark for Evolutionary LLM Kernel Search on Apple Silicon
arXiv:2605. 09708v2 Announce Type: replace-cross Abstract: We present Metal-Sci, a 10-task benchmark of scientific Apple Silicon Metal compute kernels spanning six optimization regimes (stencils, all-pairs in $n$-body problems, multi-field Boltzmann, neighbor-list molecular dynamics, multi-kernel PDE, FFT).
arXiv Machine Learning
Jul 31
RLPF: Reinforcement Learning from Performance Feedback for Code Generation
arXiv:2607. 27271v1 Announce Type: new Abstract: Code models are increasingly trained with execution feedback, but most training signals still stop at correctness.
Hugging Face Trending Papers
Sep 10
GPU-CFR: 80x Faster Counterfactual Regret Minimization by Compiling the Game to Static Dataflow and CUDA Graph Replay
GPU-CFR is a compiler and runtime that transforms any counterfactual regret minimization (CFR) game into a static dataflow representation, eliminating variable kernel launches by precomputing indices, flat arrays, and depth‑level execution blocks. This approach reduces framework operations by up to 18.1× and allows a single CUDA Graph Replay to execute each iteration, yielding 29.8–80.4× speedups over the fastest prior GPU CFR on an A100 and 14–258× over the LiteEFG CPU implementation for large games. The compiled representation alone delivers 2.2–51.1× acceleration on eight CPU threads, while the optimized path reproduces reference iterates exactly and pays for its overhead within the first solve.
arXiv Machine Learning
Sep 18
Sample Count Is Not Enough: Candidate-Generation Strategy Shapes the Energy and Performance of LLM Test-Time Scaling
The paper demonstrates that the number of candidates generated during test-time scaling of large language models does not fully capture the system cost. By comparing different generation schedules (e.g., one batched call versus multiple serial calls) while keeping the total candidate count fixed, the authors show that serial calls consume significantly more GPU energy and latency. The study suggests that reporting candidate count alone is insufficient; evaluations should also include generation schedule and GPU-level metrics.
arXiv AI
Sep 12
GPU-CFR: 80x Faster Counterfactual Regret Minimization by Compiling the Game to Static Dataflow and CUDA Graph Replay
GPU-CFR compiles a fixed game into static dataflow, eliminating per-iteration kernel launches and reducing framework operations by up to 18.1×. On an A100 GPU it achieves 29.8–80.4× speedups over the fastest prior GPU CFR and 14–258× over the LiteEFG CPU implementation for large games. The compiled representation alone delivers 2.2–51.1× acceleration on eight CPU threads, while the CUDA Graph Replay enables a single graph launch per iteration.
arXiv AI
Aug 18
StateM: Reaching 95.3% Raw Accuracy, or a \$15 Frontier Run, on Terminal-Bench 2.1 via Harness Scaling
arXiv:2608. 15089v1 Announce Type: new Abstract: Long-horizon agents can fail even when their underlying models can solve the constituent steps.
arXiv AI
Jun 30
Knowing in Advance When an Evolutionary Outer Loop Will Not Help: A Pre-Registered Cheap-Baseline Screening Rule
arXiv:2606. 29119v1 Announce Type: cross Abstract: We introduce a pre-registered screening rule that decides, before any implementation, whether an evolutionary / population / lifecycle outer loop over neural-network parameters or structure is worth building.
Hugging Face Trending Papers
Aug 20
Daedalus-150M: A Convolution-Attention Hybrid Designed for CPU Inference
Small language models are usually built like large ones and then squeezed onto a CPU afterwards. We did the opposite: we fixed the target first, one user, one token at a time, 4-bit weights, ordinary CPU, and chose the architecture to suit it.
arXiv AI
Jun 24
BluTrain: A C++/CUDA Framework for AI Systems
arXiv:2606. 24780v1 Announce Type: new Abstract: Progress in deep learning is, at scale, more a matter of systems engineering than of modelling: the behaviour of a model in training (its throughput, its memory footprint, and the numerical fidelity of the result) is determined less by the architecture itself than by how that architecture is expressed on the hardware.