DLR creates discrete latent tokens from rendered CoT images via clustering, enabling up to 20x compression and interpretable trajectories that outperform continuous latent baselines on reasoning tasks.
How do latent reasoning methods perform under weak and strong supervision?arXiv preprint arXiv:2602.22441, 2026
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Why Struggle with Continuous Latents? Interpretable Discrete Latent Reasoning via Rendered Compression
DLR creates discrete latent tokens from rendered CoT images via clustering, enabling up to 20x compression and interpretable trajectories that outperform continuous latent baselines on reasoning tasks.