AI Paper Atlas

Research map / Time series / Anomaly detection

AnomSeer: Reinforcing Multimodal LLMs to Reason for Time-Series Anomaly Detection

Junru Zhang, Lang Feng, Haoran Shi, Xu Guo, Han Yu, Yabo Dong, Duanqing Xu

ICML 2026 · Time series › Anomaly detection · approach: Time series anomaly detection evaluation · 0 citations (OpenAlex)

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
use ExpCoT (classical-analysis expert traces) instead of GPT-4o generated CoT as reasoning reference (ablation)AnomSeer-3B · AnomLLM frequency anomaliesAffinity-F1 (%): 49.8 -> 58.9Improved
use ExpCoT instead of GPT-4o CoT (ablation)AnomSeer-3B · AnomLLM trend anomaliesAffinity-F1 (%): 79.5 -> 84.9Improved
use ExpCoT instead of GPT-4o CoT (ablation)AnomSeer-3B · AnomLLM range/point anomaliesAffinity-F1 (%): range 84.4 -> 85.6; point 86.1 -> 87.8Improved
add orthogonal projection of the reasoning advantage (ablation)AnomSeer-3B · AnomLLM frequency anomaliesAffinity-F1 (%): 53.5 -> 58.9Improved
add orthogonal projection of the reasoning advantage (ablation)AnomSeer-3B · AnomLLM trend/range/point anomaliesAffinity-F1 (%): trend 81.1 -> 84.9; range 83.5 -> 85.6; point 85.4 -> 87.8Improved
add ExpCoT, time-series reasoning advantage and orthogonalization on top of vanilla GRPO (ablation)GRPO on Qwen2.5-VL-3B · AnomLLM point anomaliesAffinity-F1 (%): 80.6 -> 87.8Improved
add TimerPO components to vanilla GRPO (ablation)GRPO on Qwen2.5-VL-3B · AnomLLM frequency anomaliesAffinity-F1 (%): 50.4 -> 58.9Improved
increase SFT data tenfold (3.2k to 32k instances)Qwen2.5-VL-3B SFT · AnomLLM testaverage Affinity-F1 (%): 32.8 -> 39.4Improved
use AnomSeer-3B RL instead of SFT on 32k instancesQwen2.5-VL-3B · AnomLLM testaverage Affinity-F1 (%): 39.4 -> 79.3Improved
scale AnomSeer from 3B to 7B backboneAnomSeer · AnomLLM testaverage Affinity-F1 (%): 79.3 -> 84.4Improved
use AnomSeer-3B instead of TimeMaster-3B (SFT+GRPO)Qwen2.5-VL-3B · AnomLLM testaverage Affinity-F1 (%): 71.9 -> 79.3Improved
vary reasoning advantage weight alpha (ablation)AnomSeer-3B · AnomLLMAffinity-F1No 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