Google's SEER adds self-evolving event reasoning to time-series forecasting, beats SOTA on six benchmarks
google · hf · 2026-10-07
Google proposes SEER (Self-Evolving Event Reasoning and Retrieval), a closed-loop framework for event-conditioned time-series forecasting:
- Motivation: real-world time series are driven by exogenous events and structural shifts, and standard retrieval-augmented approaches struggle with high noise, missing signals, and lack of causal reasoning about event impacts.
- Method: translates prediction errors into two decoupled feedback mechanisms—(i) a reflective retrieval memory that refines search queries and filters spurious noise, and (ii) a persistent causal knowledge base that distills transferable domain dynamics. Strict chronological boundaries across retrieval and reflection prevent look-ahead bias and data leakage.
- Results: consistently outperforms state-of-the-art time-series foundation models and LLM baselines across six volatile benchmarks.
More from Research
- Ethan Mollick: AI will spark two science revolutions — re-reading all literature, then speeding up discoveries — emollick · 2026-10-07
- OpenAI's one-tape TM simulation implies RAM time t is in SPACE[t^4/5], says Williams — rrwilliams · 2026-10-07
- What If AI Agents Remembered Like Living Systems? A Mycelial Framework for Agent Memory — repligate · 2026-10-07
- U. Tokyo's Kavli IPMU Hires Postdocs to Build Agentic AI for Theoretical Physics — fatihdin4en · 2026-10-07
- ACL 2027 launches special theme track on LLM homogenization and knowledge collapse — TuhinChakr · 2026-10-07
- Microsoft's PrisMem evolves agent memory per-capability, beats baselines by 10.5 points on BEAM-1M — microsoft · 2026-10-07