Sycophancy Suppression Can Impair Rational Updating: Anti-Sycophancy Should Preserve the Ability to Update

Huanhuan Ma; Henry Peng Zou; Chengze Li; Enze Ma; Yunyue Su; Philip S. Yu

cs.CLAug 27, 2026

Sycophancy Suppression Can Impair Rational Updating: Anti-Sycophancy Should Preserve the Ability to Update

Authors: Huanhuan Ma, Henry Peng Zou, Chengze Li, Enze Ma, Yunyue Su, Philip S. Yu

Organizations: University of Illinois Chicago · National University of Singapore

Abstract

Large language models often exhibit sycophancy, revising their answers to align with users when users push back. Such answer flips, however, can arise from different causes. One possibility is that the model simply aligns with the user's feedback in order to satisfy them. Another is that the feedback genuinely contains useful evidence, prompting the model to update its answer in a rational way. We distinguish them as Unsupported-Yielding and Rational-Updating. Prior work focuses primarily on suppressing Unsupported-Yielding, while overlooking its effect on Rational-Updating. We address this gap with a two-turn evaluation framework that measures the two behaviors separately. Across representative training-time and inference-time interventions, we find that anti-sycophancy methods often encounter a trade-off in which reducing Unsupported-Yielding can sacrifice Rational-Updating, and vice versa, even when the two objectives are optimized jointly. Mechanistic analysis suggests that the two behaviors share an internal substrate: the MLP neurons and attention heads driving them overlap substantially, and their associated steering directions are positively aligned. We further conduct a preliminary orthogonalized steering exploration, which yields modest, backbone-dependent selectivity gains. Overall, our results suggest that anti-sycophancy should be treated not as a simple suppression problem, but as a selectivity problem, where effective interventions should preserve Rational-Updating while reducing Unsupported-Yielding.

Figures & tables

Panel of Figure 1 — Figure 1: (a) Unsupported-yielding : the model abandons a correct answer under user pushback. (b) Rational-updating : the model corrects a wrong answer once supporting evidence is provided.

Backbone	Dataset	ACC	$R_{\mathrm{UY}}$	$R_{\mathrm{RU}}^{\textsc{E}}$	$R_{\mathrm{RU}}^{\textsc{UE}}$
Llama-3.1-8B	TruthfulQA	44.4	43.7	15.6	19.3
	PopQA	35.1	60.1	67.6	62.2
	EX-FEVER	62.6	91.2	79.9	56.4
	AQuA	61.0	87.1	79.8	77.8
	Avg	50.8	70.5	60.7	53.9
Llama-3.2-3B	TruthfulQA	39.7	51.6	17.1	18.2

Table 1: Baseline rates on the calibration split (%). ACC is single-turn accuracy without feedback;

R_{\mathrm{UY}}

is the Unsupported-Yielding rate (lower better);

R_{\mathrm{RU}}^{\textsc{E}}

and

R_{\mathrm{RU}}^{\textsc{UE}}

are the Rational-Updating rates under Evidence and user-framed User-Evidence (higher better).

TruthfulQA

PopQA

EX-FEVER

AQuA

Setting

\Delta R_{\mathrm{UY}}

\Delta R_{\mathrm{RU}}^{\textsc{E}}

\Delta R_{\mathrm{RU}}^{\textsc{UE}}

\Delta R_{\mathrm{UY}}

\Delta R_{\mathrm{RU}}^{\textsc{E}}

\Delta R_{\mathrm{RU}}^{\textsc{UE}}

\Delta R_{\mathrm{UY}}

\Delta R_{\mathrm{RU}}^{\textsc{E}}

\Delta R_{\mathrm{RU}}^{\textsc{UE}}

\Delta R_{\mathrm{UY}}

\Delta R_{\mathrm{RU}}^{\textsc{E}}

\Delta R_{\mathrm{RU}}^{\textsc{UE}}

Succ.

Trade-off

Llama-3.1-8B

Anti-pressure

-26.6

+0.1

-3.0

-41.9

+9.6

+11.3

-32.9

-53.7

-48.9

-25.9

-18.3

-53.2

4/4

3/4

Rational-updating

+20.7

+7.8

+7.7

-26.4

+11.5

+18.8

+4.2

-12.5

+9.1

+11.7

+15.0

-3.0

4/4

3/4

Joint

-20.8

+11.9

+11.9

-46.9

+13.5

+20.1

-26.4

+0.9

+9.7

+11.1

+7.3

-0.2

3/4

1/4

\Delta

acc (A/R/J)

+5.0/-2.5/+0.0

+7.8/+3.7/+5.6

-7.1/+0.0/+2.6

-5.7/-0.4/+0.4

Llama-3.2-3B

Anti-pressure

-21.3

+6.9

+5.6

-28.2

+5.0

-1.2

-73.3

-18.1

-11.9

-25.7

-18.7

-21.8

4/4

3/4

Table 2: Training-time intervention performance on the test split, shown as percentage-point changes from the base model. Lower

\Delta R_{\mathrm{UY}}

and higher

\Delta R_{\mathrm{RU}}

are better. Red marks trade-off cells; the last column counts trade-off datasets out of four. Gray marks settings that did not achieve their own objective; these are counted as failures, not trade-offs. Succ. counts datasets on which the objective was achieved (

\Delta R_{\mathrm{UY}}<0

and at least one positive

\Delta R_{\mathrm{RU}}

for Joint ).

\Delta

acc : single-turn accuracy change for Anti-pressure/Rational-updating/Joint.

Panel of Figure 2 — Figure 2: Mechanistic-analysis pipeline. Unsupported-pressure and supporting-evidence feedback define $\mathcal{S}_{\mathrm{UY}}$ and $\mathcal{S}_{\mathrm{RU}}$ ; attribution and contrastive averaging then identify the component sets $V$ and steering directions $v$ for each behavior.

Figure 3: Cross-patch validation. The curve shows the recovered fraction of the prompt-induced metric shift as the number of patched components increases. Red: top-

k

components; orange: Random-LM ; gray: Random-U ; bands are

\pm 1

std over 3 seeds.

Figure 4: Per-layer distribution of the top-200 MLP neurons on TruthfulQA. Red: only

V_{\mathrm{yielding}}

; purple: overlap; green: only

V_{\mathrm{updating}}

Figure 5: Layer-wise cosine similarity between the yielding and updating steering directions on TruthfulQA. A value of

1

means the directions are identical;

0

means they are orthogonal.

Model	Sel.	Best setting	$\Delta R_{\mathrm{UY}}$	$\Delta R_{\mathrm{RU}}^{\textsc{E}}$	$\Delta R_{\mathrm{RU}}^{\textsc{UE}}$
Llama-3.1	$2\rightarrow 1$	Head, orth.	$\mathbf{-9.4}$	$\mathbf{+4.5}$	$\mathbf{+1.5}$
Llama-3.2	$0\rightarrow 3$	Layer, orth.	$\mathbf{-4.4}$	$\mathbf{+1.3}$	$\mathbf{+1.3}$
Gemma-3	$3\rightarrow 5$	Head, non-orth.	$\mathbf{-10.3}$	$\mathbf{+6.2}$	$\mathbf{+9.9}$
Qwen3	$0\rightarrow 1$	MLP, orth.	$\mathbf{+0.0}$	$\mathbf{+1.3}$	$\mathbf{+1.3}$

Table 3: TruthfulQA steering results. “Sel.” counts selective settings before/after orthogonalization. Other columns show a representative best setting, with percentage-point changes from the base model.

Appendix figures & tables9 assets

Supplementary material from the paper’s appendix.

Appendix

Dataset	Source	Task	Sum	cal	test	Evidence $e$
TruthfulQA ( Lin et al., 2022 )	truthful_qa (mc+gen)	MC1, 4-choice	$604$	$484$	$120$	Wikipedia extract
PopQA ( Mallen et al., 2023 )	akariasai/PopQA	open-ended entity QA	$2{,}000$	$1{,}000$	$1{,}000$	Wikipedia lead extract
EX-FEVER ( Ma et al., 2024 )	dependentsign/EX-FEVER	binary True/False	$2{,}000$	$1{,}000$	$1{,}000$	dataset explanation
AQuA ( Ling et al., 2017 )	deepmind/aqua_rat	MC, 5-choice (A–E)	$501$	$254$	$247$	dataset rationale
Total	–	–	$5{,}105$	$2{,}738$	$2{,}367$	–

Appendix

Table 4: Diagnostic datasets, sizes, calibration/test splits, and evidence sources. EX-FEVER drops NEI claims; AQuA has

247

test instances after option-schema filtering.

Dataset	Sample $(x,y^{\ast},e)$
TruthfulQA	Q: What is the smallest country in the world that is at least one square mile in area?
	$y^{\ast}$ : Nauru.
	$e$ : Nauru has an area of only 21 square kilometres (8.1 sq mi), which makes it the third-smallest country in the world. Monaco, the second smallest sovereign state in the world after Vatican City, has an area of 2.08 square kilometres (0.80 sq mi).
PopQA	Q: What is Magdeburg the capital of?
	$y^{\ast}$ : Saxony-Anhalt.
	$e$ : Magdeburg is the capital of the German state of Saxony-Anhalt. The city is on the Elbe river.

Appendix

Table 5: One sample

(x,y^{\ast},e)

per dataset.

Intervention	Optimization	A-p	R-u	Joint
DPO	pairs	$9/12$	$7/13$	$7/16$
SFT-on-chosen	chosen response only	$9/13$	$9/14$	$6/16$
Steering	none (training-free)	$3/8$	$5/11$	$2/12$

Appendix

Table 6: The trade-off is not specific to DPO. Trade-offs / settings that achieved their objective, under A nti- p ressure, R ational- u pdating, and Joint objectives. DPO and SFT:

16

backbone

\times

dataset settings; steering: the non-orthogonal TruthfulQA sweep (

12

settings per objective).

TruthfulQA

PopQA

EX-FEVER

AQuA

Setting

\Delta R_{\mathrm{UY}}

\Delta R_{\mathrm{RU}}^{\textsc{E}}

\Delta R_{\mathrm{RU}}^{\textsc{UE}}

\Delta R_{\mathrm{UY}}

\Delta R_{\mathrm{RU}}^{\textsc{E}}

\Delta R_{\mathrm{RU}}^{\textsc{UE}}

\Delta R_{\mathrm{UY}}

\Delta R_{\mathrm{RU}}^{\textsc{E}}

\Delta R_{\mathrm{RU}}^{\textsc{UE}}

\Delta R_{\mathrm{UY}}

\Delta R_{\mathrm{RU}}^{\textsc{E}}

\Delta R_{\mathrm{RU}}^{\textsc{UE}}

Succ.

Trade-off

Llama-3.1-8B

Anti-pressure

-34.4

-1.7

-1.8

-29.7

+1.3

+0.7

-10.5

-9.5

-9.9

+5.5

+1.1

-6.0

3/4

2/4

Rational-updating

+15.2

+12.0

+6.6

+5.6

+3.4

+4.7

+0.0

-3.4

+5.6

+9.2

-0.3

-1.3

3/4

3/4

Joint

-33.4

+9.0

+3.7

-22.9

+5.7

+6.4

-5.1

-2.8

+1.5

+6.7

+3.6

-7.2

3/4

2/4

\Delta

acc (A/R/J)

-5.8/-8.3/-9.2

-1.0/-1.5/-0.3

-14.0/-3.2/-2.1

-7.3/+0.4/+1.2

Llama-3.2-3B

Anti-pressure

-22.5

-1.8

-7.1

-21.7

-2.8

+2.1

-17.6

-30.4

-13.9

+0.8

-14.7

-8.6

3/4

3/4

Appendix

Table 7: SFT-on-chosen results on the held-out test split, as percentage-point changes from the base model. As in Table 2 , red marks trade-off cells, gray marks settings that did not achieve their objective, and Succ. counts datasets on which the objective was achieved.

Dataset	Llama-3.1	Llama-3.2	Gemma-3	Qwen3
TruthfulQA	70 / 44	78 / 42	90 / 59	64 / 34
PopQA	68 / 43	62 / 38	74 / 47	38 / 26
EX-FEVER	78 / 71	68 / 69	86 / 78	70 / 65
AQuA	80 / 66	82 / 64	86 / 80	74 / 65

Appendix

Table 8: MLP-neuron overlap between

V_{\mathrm{yielding}}

and

V_{\mathrm{updating}}

across all four datasets, reported as

k{=}50

k{=}5000

(%). Two independent uniform top-

k

selections would overlap by

k^{2}/N

, i.e. below

0.03\%

k{=}50

and

1

–

2\%

k{=}5000

for these backbones.

Backbone	TruthfulQA	PopQA	EX-FEVER	AQuA
Llama-3.1	0.58	0.42	0.58	0.61
Llama-3.2	0.54	0.49	0.59	0.62
Gemma-3	0.71	0.40	0.55	0.68
Qwen3	0.66	0.58	0.67	0.84

Appendix

Table 9:

\cos(v_{\mathrm{yielding}},v_{\mathrm{updating}})

averaged over residual-stream layers, estimated on the calibration split. All

16

backbone

\times

dataset entries are positive.

$k$	Llama-3.1	Llama-3.2	Gemma-3	Qwen3
50	70	78	90	64
100	60	61	80	54
200	52	51	72	48
500	48	43	69	43
1000	45	42	64	38
2000	45	41	60	36

Appendix

Table 10: TruthfulQA MLP-neuron overlap between

V_{\mathrm{yielding}}

and

V_{\mathrm{updating}}

k

varies (%).

Figure 6: Per-layer distributions of the top-200 MLP neurons on the other three datasets. Red: only

V_{\mathrm{yielding}}

; purple: overlap; green: only

V_{\mathrm{updating}}

Model	Locus	Objective	Non-orthogonal				Orthogonalized
			$\Delta R_{\mathrm{UY}}$	$\Delta R_{\mathrm{RU}}^{\textsc{E}}$	$\Delta R_{\mathrm{RU}}^{\textsc{UE}}$	Sel.	$\Delta R_{\mathrm{UY}}$	$\Delta R_{\mathrm{RU}}^{\textsc{E}}$	$\Delta R_{\mathrm{RU}}^{\textsc{UE}}$	Sel.
Llama-3.1	Layer	Yielding-only	$-5.7$	$+0.0$	$+0.0$		$-3.8$	$+0.0$	$+1.5$
Llama-3.1	Layer	Updating-only	$+3.8$	$+3.0$	$+1.5$		$+1.9$	$+4.5$	$+1.5$
Llama-3.1	Layer	Joint	$-1.9$	$+0.0$	$+0.0$		$-1.9$	$+0.0$	$+0.0$
Llama-3.1	Head	Yielding-only	$-3.8$	$+3.0$	$+3.0$	✓	$-3.8$	$+3.0$	$+0.0$
Llama-3.1	Head	Updating-only	$-5.7$	$+4.5$	$+0.0$		$-9.4$	$+4.5$	$+1.5$	✓

Appendix

Table 11: Full TruthfulQA steering results for § 6 . Values are percentage-point changes from the base model; bold entries and green checks mark selective settings.

LLM Sycophancy

Large language model sycophancy is the tendency to mirror users’ beliefs, preferences, or authority cues at the expense of independent judgment.

See all 151 papers

SWAY: A Counterfactual Computational Linguistic Approach to Measuring and Mitigating Sycophancy
Apr 2, 2026Joy Bhalla, Kristina GligorićLLM Sycophancy Prompt Engineering
Johns Hopkins University
Ask don't tell: Reducing sycophancy in large language models
Feb 27, 2026Magda Dubois, Cozmin Ududec, Christopher Summerfield +1LLM Sycophancy Framing Effects in Language Models
UK AI Security Institute, London, UK
Not Your Typical Sycophant: The Elusive Nature of Sycophancy in Large Language Models
Jan 21, 2026Shahar Ben-Natan, Oren TsurLLM Evaluation LLM Sycophancy
Computer and Information Science Ben Gurion University

Apr 2, 2026J/K move · Enter open · S save

SWAY: A Counterfactual Computational Linguistic Approach to Measuring and Mitigating Sycophancy

Joy Bhalla, Kristina Gligorić

Johns Hopkins University

LLM Sycophancy Prompt Engineering

Open paper AskPDF arXiv

Large language models exhibit sycophancy: the tendency to shift outputs toward user-expressed stances, regardless of correctness or consistency. While prior work has studied this issue and its impacts, rigorous computational linguistic metrics are needed to identify when models are being sycophantic. Here, we introduce SWAY, an unsupervised computational linguistic measure of sycophancy. We develop a counterfactual prompting mechanism to identify how much a model's agreement shifts under positive versus negative linguistic pressure, isolating framing effects from content. Applying this metric to benchmark 6 models, we find that sycophancy increases with epistemic commitment. Leveraging our metric, we introduce a counterfactual mitigation strategy teaching models to consider what the answer would be if opposite assumptions were suggested. While baseline mitigation instructing to be explicitly anti-sycophantic yields moderate reductions, and can backfire, our counterfactual CoT mitigation drives sycophancy to near zero across models, commitment levels, and clause types, while not suppressing responsiveness to genuine evidence. Overall, we contribute a metric for benchmarking sycophancy and a mitigation informed by it.

Abstract

LLM Sycophancy

Explore similar work

SWAY: A Counterfactual Computational Linguistic Approach to Measuring and Mitigating Sycophancy

Ask don't tell: Reducing sycophancy in large language models

Not Your Typical Sycophant: The Elusive Nature of Sycophancy in Large Language Models