SanSi: Looped Typed Decision Models for System 1.5 Thinking

This episode examines SanSi, a small "typed decision" model that returns a probability for each caller-declared option (such as spam or not spam) in a single forward pass, with no generated text to parse. It explains how the model gets more computation by looping the same 24-layer stack up to eight times, with a readout after every loop, so one model can serve any budget from one loop to eight. The discussion covers how this "System 1.5" approach differs from one-pass classifiers and from generated chain-of-thought. It also covers the earlier work it builds on: Universal Transformers, adaptive computation time, and pre-trained looped models such as Ouro. On training, the model uses a frozen backbone with LoRA adapters (about 61 million trainable parameters) and a loss averaged across all loops. Supervising only the final loop dropped loop-1 accuracy to 35.6%, against 58.4% with per-loop training. The episode also raises caveats: the Kahneman-style labels are framing rather than evidence, the RL case study uses the judge's own probability as reward rather than a truly verifiable one, and it is unclear whether looping beats spending the same compute another way. Sources: 1. SanSi: A Looped Typed Decision Model for System 1.5 Thinking — Shuyu Gan, Young-Jun Lee, Dongyeop Kang, 2026 http://arxiv.org/abs/2610.07730 2. SanSi: A Looped Typed Decision Model for System 1.5 Thinking — Shuyu Gan, Young-Jun Lee, Dongyeop Kang, 2026 https://scholar.google.com/scholar?q=SanSi%3A+A+Looped+Typed+Decision+Model+for+System+1.5+Thinking 3. Strictly Proper Scoring Rules, Prediction, and Estimation — Tilmann Gneiting, Adrian E. Raftery, 2007 https://scholar.google.com/scholar?q=Strictly+Proper+Scoring+Rules%2C+Prediction%2C+and+Estimation 4. On Calibration of Modern Neural Networks — Chuan Guo, Geoff Pleiss, Yu Sun, Kilian Q. Weinberger, 2017 https://scholar.google.com/scholar?q=On+Calibration+of+Modern+Neural+Networks 5. Generative Verifiers: Reward Modeling as Next-Token Prediction — Lunjun Zhang, Arian Hosseini, Hritik Bansal, Mehran Kazemi, Aviral Kumar, Rishabh Agarwal, 2024 https://scholar.google.com/scholar?q=Generative+Verifiers%3A+Reward+Modeling+as+Next-Token+Prediction 6. Universal Transformers — Mostafa Dehghani, Stephan Gouws, Oriol Vinyals, Jakob Uszkoreit, Łukasz Kaiser, 2019 https://scholar.google.com/scholar?q=Universal+Transformers 7. Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach — Jonas Geiping, Sean McLeish, Neel Jain, John Kirchenbauer, Siddharth Singh, Brian R. Bartoldson, Bhavya Kailkhura, Abhinav Bhatele, Tom Goldstein, 2025 https://scholar.google.com/scholar?q=Scaling+up+Test-Time+Compute+with+Latent+Reasoning%3A+A+Recurrent+Depth+Approach 8. Scaling Latent Reasoning via Looped Language Models (Ouro) — Rui-Jie Zhu et al., 2025 https://scholar.google.com/scholar?q=Scaling+Latent+Reasoning+via+Looped+Language+Models+%28Ouro%29 9. Adaptive Computation Time for Recurrent Neural Networks — Alex Graves, 2016 https://scholar.google.com/scholar?q=Adaptive+Computation+Time+for+Recurrent+Neural+Networks 10. Multi-Scale Dense Networks for Resource Efficient Image Classification — Gao Huang, Danlu Chen, Tianhong Li, Felix Wu, Laurens van der Maaten, Kilian Q. Weinberger, 2018 https://scholar.google.com/scholar?q=Multi-Scale+Dense+Networks+for+Resource+Efficient+Image+Classification 11. Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach (Huginn) — Jonas Geiping et al., 2025 https://scholar.google.com/scholar?q=Scaling+up+Test-Time+Compute+with+Latent+Reasoning%3A+A+Recurrent+Depth+Approach+%28Huginn%29 12. PonderNet: Learning to Ponder — Andrea Banino, Jan Balaguer, Charles Blundell, 2021 https://scholar.google.com/scholar?q=PonderNet%3A+Learning+to+Ponder 13. Shallow-Deep Networks: Understanding and Mitigating Network Overthinking — Yigitcan Kaya, Sanghyun Hong, Tudor Dumitras, 2019 https://scholar.google.com/scholar?q=Shallow-Deep+Networks%3A+Understanding+and+Mitigating+Network+Overthinking 14. End-to-end Algorithm Synthesis with Recurrent Networks: Logical Extrapolation Without Overthinking — Arpit Bansal, Avi Schwarzschild, Eitan Borgnia, Zeyad Emam, Furong Huang, Micah Goldblum, Tom Goldstein, 2022 https://scholar.google.com/scholar?q=End-to-end+Algorithm+Synthesis+with+Recurrent+Networks%3A+Logical+Extrapolation+Without+Overthinking 15. Proper Scoring Rules / Strictly Proper Scoring Rules, Prediction, and Estimation — Tilmann Gneiting, Adrian Raftery, 2007 https://scholar.google.com/scholar?q=Proper+Scoring+Rules+%2F+Strictly+Proper+Scoring+Rules%2C+Prediction%2C+and+Estimation 16. Large Language Models Can Self-Improve / TTRL: Test-Time Reinforcement Learning — Yuxin Zuo et al. (TTRL); Jiaxin Huang et al., 2025 https://scholar.google.com/scholar?q=Large+Language+Models+Can+Self-Improve+%2F+TTRL%3A+Test-Time+Reinforcement+Learning 17. Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena — Lianmin Zheng et al., 2023 https://scholar.google.com/scholar?q=Judging+LLM-as-a-Judge+with+MT-Bench+and+Chatbot+Arena 18. Chain-of-Thought Reasoning Without Prompting / Faithfulness work on CoT (Turpin et al., Language Models Don't Always Say What They Think) — Miles Turpin et al., 2023 https://scholar.google.com/scholar?q=Chain-of-Thought+Reasoning+Without+Prompting+%2F+Faithfulness+work+on+CoT+%28Turpin+et+al.%2C+Language+Models+Don%27t+Always+Say+What+They+Think%29 Interactive Visualization: SanSi: Looped Typed Decision Models for System 1.5 Thinking