arXiv AI
Lattice Deduction Transformers
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
Related stories
arXiv AI
Jun 16
Recurrent Reasoning on Symbolic Puzzles with Sequence Models
arXiv:2606. 15686v1 Announce Type: new Abstract: Large language models often appear strong on symbolic and algorithmic tasks, yet this apparent strength can hide brittle behaviour when problems become longer, harder, or slightly out of distribution.
arXiv AI
Oct 15
Tiny Recursive Message Passing
Tiny Recursive Message Passing applies recursive transformers and deep supervision to message passing neural networks, achieving high efficiency. The approach yields about 99% test accuracy on Sudoku-Extreme with only 16k parameters, training quickly on modern GPUs or CPUs. It also demonstrates zero‑shot transfer to larger problem instances, outperforming popular neural baselines on N‑queens and Boolean satisfiability with fewer parameters.
arXiv AI
Jul 23
Anatomy of a Sound Neural Reasoner: One-Shot Amortization, First-Pass Poisoning, and Search Inertness in Clue-Rich Completion
arXiv:2607. 19635v1 Announce Type: cross Abstract: Neural solvers are built to deduce, branch, and revise intermediate states.
Hugging Face Trending Papers
Jul 22
Anatomy of a Sound Neural Reasoner: One-Shot Amortization, First-Pass Poisoning, and Search Inertness in Clue-Rich Completion
Neural solvers are built to deduce, branch, and revise intermediate states. The Lattice Deduction Transformer (LDT) appears to do exactly that.
arXiv Machine Learning
Oct 28
Resample the Answer: Where to Spend Test-Time Compute in Recursive Reasoning
The paper investigates how to allocate test‑time computation in recursive reasoning models, specifically the Tiny Recursive Model (TRM). Experiments show that resampling the answer state (y) is more effective than perturbing the reasoning state (z) for recovering failures, especially as compute budgets increase. The proposed Gated Answer Resampling (GAR) method, which restarts only rejected puzzles, boosts TRM performance from 85.9% to 99.4% on Sudoku‑Extreme and from 68.8% to 85.6% on the Pencil Puzzle Bench, outperforming the probabilistic variant that uses far more compute.
arXiv AI
Jun 30
Flow Reasoning Models: Scaling Reasoning Through Iterative Self-Refinement
arXiv:2606. 29150v1 Announce Type: new Abstract: Discrete flow models have recently shown promising performance on few-step text generation; however, when naively applied to structured reasoning tasks such as Sudoku and Zebra puzzles, they converge confidently to incorrect answers (solving only $\sim$36% of Sudoku puzzles).