@inproceedings{singh-kaynig-fittkau-2026-decomposing,
title = "Decomposing {LLM} Advice Sycophancy: Content, Style, and Framing Drive Distinct Response Dimensions",
author = "Singh, Jaya and
Kaynig-Fittkau, Verena",
editor = "Zhang, Yue and
Xu, Ruifeng and
Purwarianti, Ayu and
Rong, Tong and
Suzuki, Jun",
booktitle = "Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the {A}sia-Pacific Chapter of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
month = nov,
year = "2026",
address = "Hengqin, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.aacl-long.12/",
pages = "255--274",
ISBN = "979-8-89176-439-2",
abstract = "Large language models are widely used for personal advice (relationship conflicts, financial decisions, mental-health support, etc.), but they often align their response with the user{'}s position rather than offering independent judgment, even when users need pushback rather than validation. Mitigating this, whether through training objectives, system-level guardrails, or evaluation design, requires knowing what triggers it. But what in the user{'}s message triggers this behavior is unknown. Natural advice-seeking messages vary simultaneously in semantic content (what is disclosed), linguistic style (how it is written), and narrative framing (whose side is told), but existing benchmarks cannot attribute model behavior to any single cue. Without this attribution, mitigations cannot be targeted, for example, suppressing warmth may harm users with genuine distress but might not fix the model{'}s tendency to endorse whichever side of a conflict speaks. In this work, we isolate which of these cues drive which aspect of sycophantic behavior across 787 advice-seeking scenarios and observe three patterns emerge: emotional warmth is content-driven; stance endorsement is framing-driven: endorsement drops by ${\sim}50${\%} when the same conflict is retold from the opposing perspective; and recommendation directness is style-driven. We also find that hedging reduces rather than increases model endorsement; models do not simply validate vulnerable-sounding users."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="singh-kaynig-fittkau-2026-decomposing">
<titleInfo>
<title>Decomposing LLM Advice Sycophancy: Content, Style, and Framing Drive Distinct Response Dimensions</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jaya</namePart>
<namePart type="family">Singh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Verena</namePart>
<namePart type="family">Kaynig-Fittkau</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yue</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruifeng</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ayu</namePart>
<namePart type="family">Purwarianti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tong</namePart>
<namePart type="family">Rong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jun</namePart>
<namePart type="family">Suzuki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Hengqin, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-439-2</identifier>
</relatedItem>
<abstract>Large language models are widely used for personal advice (relationship conflicts, financial decisions, mental-health support, etc.), but they often align their response with the user’s position rather than offering independent judgment, even when users need pushback rather than validation. Mitigating this, whether through training objectives, system-level guardrails, or evaluation design, requires knowing what triggers it. But what in the user’s message triggers this behavior is unknown. Natural advice-seeking messages vary simultaneously in semantic content (what is disclosed), linguistic style (how it is written), and narrative framing (whose side is told), but existing benchmarks cannot attribute model behavior to any single cue. Without this attribution, mitigations cannot be targeted, for example, suppressing warmth may harm users with genuine distress but might not fix the model’s tendency to endorse whichever side of a conflict speaks. In this work, we isolate which of these cues drive which aspect of sycophantic behavior across 787 advice-seeking scenarios and observe three patterns emerge: emotional warmth is content-driven; stance endorsement is framing-driven: endorsement drops by \sim50% when the same conflict is retold from the opposing perspective; and recommendation directness is style-driven. We also find that hedging reduces rather than increases model endorsement; models do not simply validate vulnerable-sounding users.</abstract>
<identifier type="citekey">singh-kaynig-fittkau-2026-decomposing</identifier>
<location>
<url>https://aclanthology.org/2026.aacl-long.12/</url>
</location>
<part>
<date>2026-11</date>
<extent unit="page">
<start>255</start>
<end>274</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Decomposing LLM Advice Sycophancy: Content, Style, and Framing Drive Distinct Response Dimensions
%A Singh, Jaya
%A Kaynig-Fittkau, Verena
%Y Zhang, Yue
%Y Xu, Ruifeng
%Y Purwarianti, Ayu
%Y Rong, Tong
%Y Suzuki, Jun
%S Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2026
%8 November
%I Association for Computational Linguistics
%C Hengqin, China
%@ 979-8-89176-439-2
%F singh-kaynig-fittkau-2026-decomposing
%X Large language models are widely used for personal advice (relationship conflicts, financial decisions, mental-health support, etc.), but they often align their response with the user’s position rather than offering independent judgment, even when users need pushback rather than validation. Mitigating this, whether through training objectives, system-level guardrails, or evaluation design, requires knowing what triggers it. But what in the user’s message triggers this behavior is unknown. Natural advice-seeking messages vary simultaneously in semantic content (what is disclosed), linguistic style (how it is written), and narrative framing (whose side is told), but existing benchmarks cannot attribute model behavior to any single cue. Without this attribution, mitigations cannot be targeted, for example, suppressing warmth may harm users with genuine distress but might not fix the model’s tendency to endorse whichever side of a conflict speaks. In this work, we isolate which of these cues drive which aspect of sycophantic behavior across 787 advice-seeking scenarios and observe three patterns emerge: emotional warmth is content-driven; stance endorsement is framing-driven: endorsement drops by \sim50% when the same conflict is retold from the opposing perspective; and recommendation directness is style-driven. We also find that hedging reduces rather than increases model endorsement; models do not simply validate vulnerable-sounding users.
%U https://aclanthology.org/2026.aacl-long.12/
%P 255-274
Markdown (Informal)
[Decomposing LLM Advice Sycophancy: Content, Style, and Framing Drive Distinct Response Dimensions](https://aclanthology.org/2026.aacl-long.12/) (Singh & Kaynig-Fittkau, AACL-IJCNLP 2026)
ACL