pith. sign in

Title resolution pending

2 Pith papers cite this work. Polarity classification is still indexing.

2 Pith papers citing it

fields

cs.AI 1 cs.CL 1

years

2025 2

verdicts

UNVERDICTED 2

representative citing papers

Thinking Sparks!: Emergent Attention Heads in Reasoning Models During Post Training

cs.AI · 2025-09-30 · unverdicted · novelty 6.0

Post-training on reasoning tasks sparks the emergence of specialized attention heads that enable structured computation, with SFT adding stable heads while GRPO uses dynamic activation and pruning tied to reward signals, and controllable think models relying on compensatory heads instead of specific

Ultra-Low-Dimensional Prompt Tuning via Random Projection

cs.CL · 2025-02-06 · unverdicted · novelty 6.0

ULPT optimizes prompts in ultra-low dimensions with frozen random up-projection to cut training parameters by 98% while matching vanilla prompt tuning performance on NLP tasks.

citing papers explorer

Showing 2 of 2 citing papers.

  • Thinking Sparks!: Emergent Attention Heads in Reasoning Models During Post Training cs.AI · 2025-09-30 · unverdicted · none · ref 44

    Post-training on reasoning tasks sparks the emergence of specialized attention heads that enable structured computation, with SFT adding stable heads while GRPO uses dynamic activation and pruning tied to reward signals, and controllable think models relying on compensatory heads instead of specific

  • Ultra-Low-Dimensional Prompt Tuning via Random Projection cs.CL · 2025-02-06 · unverdicted · none · ref 59

    ULPT optimizes prompts in ultra-low dimensions with frozen random up-projection to cut training parameters by 98% while matching vanilla prompt tuning performance on NLP tasks.