Paper

Dual Hierarchical Dialogue Policy Learning for Legal Inquisitive Conversational Agents

Unreadunread

AI summary

arXiv:2605. 14057v1 Announce Type: new Abstract: Most existing dialogue systems are user-driven, primarily designed to fulfill user requests.

Reading status

Priority

About this paper

Authors: Xubo Lin, Zezhii Deng, Shihao Wang, Grace Hui Yang, Yang Deng
Released: 2026-05-15 (3mo ago)
Topic: Related to current project
Relation to my work: Read next because Dual Hierarchical Dialogue Policy Learning for Legal Inquisitive Conversational Agents overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)", clean result "Coupling evil personas with wrong answers fails to protect Qwen2.5-7B from EM-induced alignment collapse — and the apparent capability ordering across coupling conditions is mostly eval contamination (LOW confidence)". Matching terms: fill, eval, line, rate. Source: arxiv cs.CL (NLP).
arXiv: 2605.14057

Ask Claude about this paper

Posts a comment with @claude so the runner spins up a Claude Code session scoped to Dual Hierarchical Dialogue Policy Learning for Legal Inquisitive Conversational Agents. Replies appear in the conversation below.

Comments

No comments yet.

Related records

No links yet.