# Beyond Supervised Clarification: Input Rewriting with LLMs for Dialogue Discourse Parsing

arXiv: https://arxiv.org/abs/2607.01964 · HTML: https://mlsanity.com/paper/2607.01964/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 18 of its 66 references that are in the map.
- First posted on arXiv 2 Jul 2026.

## Builds on (most cited first, 18 of 18)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [BERTScore: Evaluating Text Generation with BERT](https://mlsanity.com/paper/1904.09675/) (2019)
- [HybridFlow: A Flexible and Efficient RLHF Framework](https://mlsanity.com/paper/2409.19256/) (2024)
- [Decontextualization: Making Sentences Stand-Alone](https://mlsanity.com/paper/2102.05169/) (2021)
- [CONQRR: Conversational Query Rewriting for Retrieval with Reinforcement Learning](https://mlsanity.com/paper/2112.08558/) (2021)
- [Molweni: A Challenge Multiparty Dialogues-based Machine Reading Comprehension Dataset with Discourse Structure](https://mlsanity.com/paper/2004.05080/) (2020)
- [A Deep Sequential Model for Discourse Parsing on Multi-Party Dialogues](https://mlsanity.com/paper/1812.00176/) (2018)
- [Rec-R1: Bridging Generative Large Language Models and User-Centric Recommendation Systems via Reinforcement Learning](https://mlsanity.com/paper/2503.24289/) (2025)
- [Improving Multi-Party Dialogue Discourse Parsing via Domain Integration](https://mlsanity.com/paper/2110.04526/) (2021)
- [Lexical Normalization for Code-switched Data and its Effect on POS-tagging](https://mlsanity.com/paper/2006.01175/) (2020)
- [Adversarial Text Normalization](https://mlsanity.com/paper/2206.04137/) (2022)
- [What Causes the Failure of Explicit to Implicit Discourse Relation Recognition?](https://mlsanity.com/paper/2404.00999/) (2024)
- [Automatic Input Rewriting Improves Translation with Large Language Models](https://mlsanity.com/paper/2502.16682/) (2025)
- [ICR: Iterative Clarification and Rewriting for Conversational Search](https://mlsanity.com/paper/2509.05100/) (2025)
- [Improving Dialogue Discourse Parsing through Discourse-aware Utterance Clarification](https://mlsanity.com/paper/2506.15081/) (2025)
- [Rethinking On-policy Optimization for Query Augmentation](https://mlsanity.com/paper/2510.17139/) (2025)
- [Towards Actionable Pedagogical Feedback: A Multi-Perspective Analysis of Mathematics Teaching and Tutoring Dialogue](https://mlsanity.com/paper/2505.07161/) (2025)

## Abstract (arXiv)

Rewriting inputs to improve frozen downstream models has become a common strategy in modern NLP pipelines. Prior work on incremental dialogue discourse parsing (DDP) shows that supervised clarification models can rewrite fragmentary or underspecified utterances, such as resolving ellipsis or references, to improve parsing accuracy. In this work, we revisit this idea under realistic deployment conditions, where no clarification supervision is available and the clarifier must rely on zero-shot prompting or feedback from a frozen parser. Across three Segmented Discourse Representation Theory (SDRT) datasets and multiple parsers, we find that last-utterance clarification is far less reliable than suggested by supervised settings. Parser-agnostic rewriting often introduces more regressions than repairs, as edits that enable fixes also disrupt discourse cues relied upon by the parser. A best-of-8 rewriting analysis further reveals a practical ceiling: a large fraction of errors are not repairable through input rewriting alone. A parser-aware clarifier trained with GRPO reduces regressions by up to 37% by learning conservative abstention, yet still fails to produce selectivity-aware clarifications that consistently improve parsing. Together, these findings recast clarification as a selective intervention problem. We identify rewritability prediction, deciding whether an utterance is repairable before intervention, as the key missing capability for input-side optimization of frozen discourse parsers, and a critical direction for improving agentic pipelines more broadly.