The Geometric Reasoner: Manifold-Informed Latent Foresight Search for Long-Context Reasoning

Ren Zhuang, Ben Wang, Shuifa SunICML 2026 · OpenReview

Abstract

Scaling test-time compute enhances long chain-of-thought (CoT) reasoning, yet existing approaches face a fundamental trade-off between computational cost and coverage quality: either incurring high training expense or yielding redundant trajectories. We introduce The Geometric Reasoner (TGR), a training-free framework that performs manifold-informed latent foresight search under strict memory bounds. At each chunk boundary, TGR scores candidate latent anchors via a lightweight look-ahead estimate combined with soft geometric regularizers that encourage smooth trajectories and diverse exploration. Chunk-wise KV cache resets keep memory linear in chunk length. On challenging math and code benchmarks, TGR improves robust trajectory coverage, measured by the area under the Pass@$k$ curve (AUC), by up to 13 points on Qwen3-8B, with negligible overhead of about 1.1--1.3$\times$.
BibTeX
@inproceedings{zhuang2026geometric,
  author = {Ren Zhuang and Ben Wang and Shuifa Sun},
  title = {{The Geometric Reasoner: Manifold-Informed Latent Foresight Search for Long-Context Reasoning}},
  booktitle = {International Conference on Machine Learning},
  year = {2026},
  url = {https://openreview.net/forum?id=mw7BL7AvQc},
}

Loading related content...

Figure