AI·llama.cpp build b9966 fixes regex recompilations for -sm tensor mode, caching patterns per tensor to reduce CPU overhead on decode threads.
Search
What are you looking for?
6 results for “optimization”
AI·LLM-based ASR in regulated domains like banking is limited by privacy and real-speech collection costs; synthetic TTS data is a cost-effective substitute, but acoustic mismatch hinders supervised fine-tuning (SFT). Group Relative Policy Optimization (GRPO) applied solely to synthetic speech reduces WER by 40% relative to SFT (36.71% to 22.09%) and by 45% in the SFT-then-GRPO sequence, by improving behavioral calibration and audio attention rather than representations.
AI·COBART uses a bidirectional auto-regressive transformer for optimized ad headline generation with multi-objective control over quality and CTR.
AI·Follow-up post details data-engineering optimizations that scaled a SQLite/FTS5 patent database from 3.5M to 5.36M records while classifying them with Nemotron 9B on an RTX 5090.
Reinforcement learning (RL) is becoming increasingly important for post-training large language models (LLMs). Previous RL pipelines for LLMs were mostly synchronous and batch-interleaved, which is inefficient for long-horizon agentic tasks. Recently, asynchronous RL has emerged as a more efficient alternative by updating the model as rollouts arrive. However, existing asynchronous RL systems often emphasize throughput, while leaving training stability and task effectiveness largely underexplor…
Reinforcement learning (RL) has become the standard paradigm for enhancing the complex reasoning capabilities of large language models (LLMs). To achieve sample efficiency, modern RL frameworks rely on importance sampling (IS). However, these algorithms suffer from an exploration-stability dilemma. Pure IS often leads to catastrophic training instability, while standard clipping mechanisms used to mitigate this instability strictly constrain the policy update budget. By formalizing the concept …