gradient-based

Tag

Cards List
#gradient-based

Is Human-Readable Text Necessary for Effective LLM Fine-Tuning?

arXiv cs.AI ↗ · 2026-09-30 Cached

This arXiv paper proposes Desired-Update-Aligned Synthetic Data (DASA), which uses activation-gradient feedback from a frozen reference model to optimize continuous synthetic input embeddings for LLM fine-tuning, questioning whether human-readable text is necessary. Experiments across Llama and Qwen models (1B–32B) show DASA matches or outperforms natural-language data under LoRA and delivers a 3.6–4.9x speedup over GRADMM.

0 favorites 0 likes
#gradient-based

Gradient-Based Speech-to-Text Alignment for Any ASR Model: From CTC to Speech LLMs

arXiv cs.CL ↗ · 2026-07-09 Cached

This paper introduces a gradient-based speech-to-text alignment method applicable to any differentiable ASR model, including CTC, transducer, attention-based encoder-decoder, and speech large language models, requiring no training or model modification.

0 favorites 0 likes
#gradient-based

GBC: Gradient-Based Connections for Optimizing Multi-Agent Systems

Hugging Face Daily Papers ↗ · 2026-06-26 Cached

Proposes Gradient-Based Connections (GBC), a method that models multi-agent LLM systems as computational graphs and uses gradient signals to attribute errors to specific agents, enabling better system-level optimization.

0 favorites 0 likes
← Back to home

Submit Feedback