AI Paper Atlas

Research map / Large language models / Alignment and preferences

AlignSurvey: A Comprehensive Benchmark for Human Preferences Alignment in Social Surveys

Chenxi Lin, Weikang Yuan, Zhuoren Jiang, Biao Huang, Ruitao Zhang, Jianan Ge, Yueqian Xu, Jianxing Yu

AAAI 2026 · Large language models › Alignment and preferences · approach: Steering and correcting LLM behavior

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
fine-tune with two-stage SurveyLM alignmentQwen2.5-7B · Task 3 stance accuracyaccuracy: 36.88% -> 56.83%Improved
fine-tune with two-stage SurveyLM alignmentMistral-7B-v0.3 · Task 4 ASEaccuracy: 28.04% -> 54.59%Improved
fine-tune Qwen2.5-7B instead of using Qwen2.5-72BQwen2.5 · Task 1accuracy: 37.83% -> 54.65%Improved
add few-shot examples (3)general LLMs · Task 1 / Task 3accuracy: GPT-4o Task1 44.08% -> 40.59%; Task3 38.12% -> 42.73%Mixed
add few-shot examplesgeneral LLMs · Task 3 group-levelWasserstein distance: GPT-4o 1.648 -> 1.180Improved
fine-tune SurveyLMQwen2.5-7B · Task 4 ASEWasserstein distance: 1.8479 -> 0.0541Improved
add Stage I foundation adaptation (ablation)SurveyLM Mistral · Task 4 ASEaccuracy: 0.53 -> 0.55Improved
add Stage II task-specific SFT (ablation)SurveyLM Mistral · Task 4 ASEaccuracy: 0.36 -> 0.55Improved
add Stage I foundation adaptation (ablation)SurveyLM Qwen · Task 4 ASEWasserstein distance: 0.889 -> 0.054Improved
SurveyLM on Task 2 naturalnessMistral-7B · Task 2naturalness score: 3.21 -> 2.61Hurt
SurveyLM Task 3 group Wasserstein vs few-shot MistralMistral-7B · Task 3 groupWasserstein distance: 0.302 -> 0.297No clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy