When Does Recurrence Become an Algorithm? Convergence Selection in Weight-Tied Looped Transformers

• URL: http://arxiv.org/abs/2607.20594v1
• Date: Wed, 22 Jul 2026 16:04:31 GMT
• Title: When Does Recurrence Become an Algorithm? Convergence Selection in Weight-Tied Looped Transformers
• Abstract summary: Weight-tied looped transformer -- one block applied T times -- implement an algorithm?<n>We answer with four findings from controlled populations on group word problems.
• Score: 17.78341274611665
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: When does a weight-tied looped transformer -- one block applied T times -- implement an actual algorithm? We answer with four findings from controlled populations on group word problems. (1) The budget law: free training installs a linear computation frontier, a mechanism that solves v positions per loop, whose speed is priced by the training contract: v ~ n_train/T_train (exponent 0.98 +/- 0.04, R^2=0.99), exactly unity under T=n training. SGD selects a frontier matching the minimum the contract demands; granting more test-time loops than ever trained rescues late positions at fixed input length, yielding a principled halting rule T* = ceil(n / v-hat). (2) Architecture prior, not expressivity, picks the algorithm: standard-depth transformers learn parallel scans on this family; weight tying flips the selection to the serial frontier, even when positional addressing for a log-depth scan is supplied. At matched depth and parameters, untied models extrapolate worst and fail to learn A5 at all. (3) The walls are not where circuit complexity says: NC1-completeness costs nothing (A5 generalizes fully), while group order does (S5's 120x120 operator deadlocks joint learning) -- and an operator-first curriculum dissolves the wall in every seed. (4) Mechanisms are portable, not mandatable: warm-starting across budget contracts transfers the algorithm in every seed, re-pricing its speed, while imposing seriality through the input schedule fails where free training succeeds. These results are invisible to standard instruments, which provably saturate at the fixed points trained loops converge to. We introduce a head instrument, the convergence-time scaling tau(n,i), validate it causally via damage cones whose slope reproduces v, and show in-distribution head measurements predict out-of-distribution fate where tail metrics do not. Results replicate on the public easy-to-hard benchmark.

Related papers

• HEAT: Faster Fully Homomorphic Inference via Approximations-Weights Co-Adaptation [56.14416807766864]
  Homomorphic encryption allows a server to run a language model directly on encrypted user prompts.<n>Existing approaches fix the iteration counts uniformly across the model rather than tailoring them to each site's error tolerance.<n>We introduce Homomorphic Encryption-Aware Training (HEAT), a fine-tuning method that makes the per-nonline iteration counts learnable.
  arXiv  Detail & Related papers  (2026-09-01T18:02:02Z)
• The Cost of Adaptivity: Matching Lower Bounds Across Learning Problems [1.0742675209112622]
  We formalize nuisance adaptation via a slice-normalized minimax ratio.<n>We also define the cost of expanding from one preannounced Gaussian query to arbitrary post-hoc inspection.
  arXiv  Detail & Related papers  (2026-08-09T17:19:11Z)
• Stabilizing Extrapolation in Looped Transformers via Learned Stochastic Stopping [56.14767235650558]
  Looped Transformers, which repeatedly apply a shared transformer block, are an architecturally natural fit for variable-length computation tasks.<n>We trace this variance to the spurious correlation in simple algorithmic tasks between sequence length and number of loops.<n>Our work suggests that "when to stop" should be treated as a training design choice, not merely an inference-time-allocation rule.
  arXiv  Detail & Related papers  (2026-06-29T08:58:09Z)
• REVES: REvision and VErification--Augmented Training for Test-Time Scaling [53.197756110943395]
  We propose a two-stage iterative framework that alternates between online data/prompt augmentation and policy optimization.<n>We observe gains of +6.5 points over the RL baseline and +4.0 points over standard multi-turn training.
  arXiv  Detail & Related papers  (2026-06-17T10:37:23Z)
• Transformers Provably Learn to Internalize Chain-of-Thought [65.41010769606844]
  Implicit Chain-of-Thought (ICoT) trains models to internalize intermediate steps within their hidden states.<n>Log-ICoT removes thinking tokens one at a time, reducing the number of stages from linear in $k$ to logarithmic.<n>Experiments on multi-layer transformers confirm the theory and visualize how reasoning is progressively absorbed into deeper layers.
  arXiv  Detail & Related papers  (2026-05-27T15:17:06Z)
• Decoupled Continuous-Time Reinforcement Learning via Hamiltonian Flow [1.8824572526199168]
  Real-world control problems evolve in continuous time with non-uniform, event-driven decisions.<n>As time gaps shrink, the $Q$-function collapses to the value function $V$, eliminating action ranking.<n>Existing continuous-time methods reintroduce action information via an advantage-rate function $q$.<n>We propose a novel decoupled continuous-time actor-critic algorithm with alternating updates.
  arXiv  Detail & Related papers  (2026-02-16T09:35:25Z)
• INC: An Indirect Neural Corrector for Auto-Regressive Hybrid PDE Solvers [61.84396402100827]
  We propose the Indirect Neural Corrector ($mathrmINC$), which integrates learned corrections into the governing equations.<n>$mathrmINC$ reduces the error amplification on the order of $t-1 + L$, where $t$ is the timestep and $L$ the Lipschitz constant.<n>We test $mathrmINC$ in extensive benchmarks, covering numerous differentiable solvers, neural backbones, and test cases ranging from a 1D chaotic system to 3D turbulence.
  arXiv  Detail & Related papers  (2025-11-16T20:14:28Z)
• Linear Transformers Implicitly Discover Unified Numerical Algorithms [28.247731784318404]
  We train a linear attention transformer on millions of masked-block matrix completion tasks.<n>The model sees only input-output pairs and a mean-squared loss; it is given no normal equations, no handcrafted iterations, and no hint that the tasks are related.<n>We prove that this rule achieves second-order convergence on full-batch problems, cuts distributed iteration complexity, and remains accurate with rank-limited attention.
  arXiv  Detail & Related papers  (2025-09-24T02:19:04Z)
• From Continual Learning to SGD and Back: Better Rates for Continual Linear Models [50.11453013647086]
  We analyze the forgetting, i.e., loss on previously seen tasks, after $k$ iterations.<n>We develop novel last-iterate upper bounds in the realizable least squares setup.<n>We prove for the first time that randomization alone, with no task repetition, can prevent catastrophic in sufficiently long task sequences.
  arXiv  Detail & Related papers  (2025-04-06T18:39:45Z)
• Intersection of Parallels as an Early Stopping Criterion [64.8387564654474]
  We propose a method to spot an early stopping point in the training iterations without the need for a validation set. For a wide range of learning rates, our method, called Cosine-Distance Criterion (CDC), leads to better generalization on average than all the methods that we compare against.
  arXiv  Detail & Related papers  (2022-08-19T19:42:41Z)
• Training Overparametrized Neural Networks in Sublinear Time [14.918404733024332]
  Deep learning comes at a tremendous computational and energy cost. We present a new and a subset of binary neural networks, as a small subset of search trees, where each corresponds to a subset of search trees (Ds) We believe this view would have further applications in analysis analysis of deep networks (Ds)
  arXiv  Detail & Related papers  (2022-08-09T02:29:42Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.