pith. sign in

Controllable preference optimization: Toward controllable multi-objective alignment

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

2 Pith papers citing it

citation-role summary

background 1

citation-polarity summary

fields

cs.AI 1 cs.CL 1

years

2026 2

verdicts

UNVERDICTED 2

roles

background 1

polarities

background 1

representative citing papers

Learning to Control Summaries with Score Ranking

cs.CL · 2026-04-19 · unverdicted · novelty 6.0

A score-ranking loss enables controllable summarization by aligning outputs to evaluation scores, matching SOTA performance with dimension-specific control on LLaMA, Qwen, and Mistral.

citing papers explorer

Showing 2 of 2 citing papers.

  • Explaining and Breaking the Safety-Helpfulness Ceiling via Preference Dimensional Expansion cs.AI · 2026-05-12 · unverdicted · none · ref 44 · 2 links

    MORA breaks the safety-helpfulness ceiling in LLMs by pre-sampling single-reward prompts and rewriting them to incorporate multi-dimensional intents, delivering 5-12.4% gains in sequential alignment and 4.6% overall improvement in simultaneous alignment.

  • Learning to Control Summaries with Score Ranking cs.CL · 2026-04-19 · unverdicted · none · ref 14

    A score-ranking loss enables controllable summarization by aligning outputs to evaluation scores, matching SOTA performance with dimension-specific control on LLaMA, Qwen, and Mistral.