This paper introduces Dark-State Invariants, a deterministic framework for identifying temporally stable regions in belief propagation dynamics. By detecting invariant-stable nodes—where both sign and magnitude remain unchanged across iterations—the method quantifies computational redundancy within iterative decoding systems. Evaluated across deterministic stress scenarios, the framework reveals that convergent systems rapidly accumulate invariant regions, while oscillatory and chaotic systems do not. This work reframes convergence as the emergence of invariance and establishes a foundation for invariant-driven computation elimination and selective update strategies in QLDPC decoding.
This paper introduces Dark-State Invariants, a deterministic framework for identifying temporally stable regions in belief propagation dynamics. By detecting invariant-stable nodes—where both sign and magnitude remain unchanged across iterations—the method quantifies computational redundancy within iterative decoding systems. Evaluated across deterministic stress scenarios, the framework reveals that convergent systems rapidly accumulate invariant regions, while oscillatory and chaotic systems do not. This work reframes convergence as the emergence of invariance and establishes a foundation for invariant-driven computation elimination and selective update strategies in QLDPC decoding.
This work introduces a deterministic framework for identifying, validating, and falsifying structural invariants in decoder transformation pipelines. Departing from traditional performance-centric analyses of belief propagation, the study treats decoder execution as a composition of deterministic operators and focuses on invariant properties that persist under repeated application. We formally establish several classes of invariants, including ternary closure, operator idempotence, and dark-skip cascade behavior, and validate them directly against implementation through deterministic testing. In parallel, we construct explicit counterexamples to demonstrate the failure of commonly assumed properties such as universal absorbing states and global idempotence under damping. A key contribution is the introduction of invariant-guided reasoning as a mechanism for eliminating redundant computation rather than optimizing it. By reframing decoder pipelines as dynamical systems with provable algebraic structure, this work provides a foundation for deterministic runtime reduction, modular pipeline design, and future extensions into qudit-based decoding and sonification-driven analysis. This work builds on prior results in dark-state invariant detection and forms part of an ongoing research program in invariant-driven computation and quantum-inspired information dynamics.
In this paper, we propose a multistage decoding framework that leverages internal information produced by an underlying message-passing decoder. The proposed method targets the failure dynamics caused by both classical trapping sets and degenerate errors supported on symmetric stabilizers, which are among the primary limitations of iterative decoding for QLDPC codes. To identify unreliable variable nodes, we introduce a heuristic metric that combines several dynamical features of the decoder, including variable-node log likelihood reliabilities, hard-decision oscillations, the number of adjacent unsatisfied checks, and the soft information contributed by unsatisfied checks. Based on this ranking metric, the decoder performs guided rewinds by selectively forcing the initial log likelihood ratio values of the most suspicious variable nodes and restarting the message-passing decoder under the corresponding forced configuration. To manage the combinatorial growth of candidate configurations, the search is formulated within a beam- search framework with controlled beam width. In addition, we introduce a pruning metric based on the combination of the residual syndrome weight and a posteriori reliability of the decoder output, thereby retaining only the most promising search paths. Logical error rate results demonstrate that the proposed decoder significantly outperforms the normalized min- sum decoder and achieves competitive performance with belief propagation enhanced by order-10 ordered statistics decoding.
This compendium presents four revised studies of deterministic computation, reusable intermediate results, and transformation invariants in belief-propagation diagnostics and quantum error-correction software. It connects earlier QSOL-IMC preprints with commit-pinned QEC source snapshots, reproducible experiments, and machine-checked mathematical models. The four studies examine baseline equivalence between min-sum decoding and unit-weight uniformly reweighted belief propagation; trace-indexed reuse of sign vectors and CRC32 values; the interaction of clipping, damping, freezing, and ternary projection in decoder transformation pipelines; and the interpretation of local “dark-state” observables in belief-propagation traces. The deposit contains four revised papers, editable manuscript sources, a dependency-free Lean 4 project, source snapshots, validation scripts, measured results, proof logs, provenance records, and integrity manifests. The formalization comprises 43 compiled theorem declarations. Empirical validation includes 576 paired decoder cases, 80 cached-versus-inline diagnostic comparisons, a reproduced reduction from 81 to 20 sign computations, 204 selected historical tests, and 18 synthetic dark-state scenario runs. The analysis also documents counterexamples and boundary conditions involving CRC32 collisions, cache provenance, source mutation, damping, NaN handling, transformation order, and the distinction between observed local stability and safe computational reuse. The Lean proofs establish properties of the stated exact mathematical models. They do not constitute a complete refinement proof for the Python/NumPy implementations or IEEE-754 arithmetic, nor do they establish physical quantum-hardware correctness or justify skipping live-decoder updates. Historical runtime improvements are identified as previously reported results, separate from the measurements reproduced in this deposit.
This work introduces QEC, a deterministic, replay-safe reasoning system for quantum error correction and invariant-driven computation. Unlike conventional adaptive systems that rely on stochastic exploration, QEC operates as a deterministic reasoning layer over system evolution, producing proof-carrying state transitions that are fully reproducible and verifiable. The system replaces probabilistic optimization with a structured pipeline of invariant detection, deterministic control, memory accumulation, governance, validation, repair reasoning, and invariant-preserving compression. Each stage produces canonical, hash-stable artifacts, ensuring that identical inputs yield byte-identical outputs across environments. QEC reframes failure as a structural phenomenon ("dark states") and introduces deterministic repair through symmetry breaking, validated via counterfactual replay. The system further demonstrates that reasoning itself can be compressed into minimal canonical forms without loss of replay fidelity. Empirical runs over QLDPC Tanner graph simulations demonstrate zero hash divergence across independent executions and compact proof artifacts (~2–5 KB), supporting the feasibility of deterministic reasoning with proof in complex systems. This work establishes a new paradigm: execution → reasoning → proof
Belief-propagation (BP) decoding for quantum low-density parity-check (QLDPC) codes is appealing due to its low complexity, yet it often exhibits convergence issues due to quantum degeneracy and short cycles that exist in the Tanner graph. To overcome this challenge, this paper proposes a reinforcement-learning (RL) approach that learns (offline) how to decode QLDPC codes based on sequential decoding trajectories. The decoding is formulated as a Markov decision process with a local, syndrome-driven state representation of the underlying RL agent. To enable fast inference, critical for practical implementation, we incrementally update our RL-based QLDPC decoder using second-order neighborhoods that avoid global rescans. Simulation results on representative QLDPC codes demonstrate the superiority of the proposed RL-based QLDPC decoders in terms of performance and convergence speed when compared to flooding and random sequential schedules, while achieving performance competitive with state-of-the-art BP-based decoders at comparable complexity.
This document presents a formally validated optimization within the QSOL-IMC Quantum Error Correction (QEC) framework (v68.5.0). The optimization exploits a data-level invariant in belief propagation (BP) dynamics diagnostics: sign vectors and their derived CRC32 signatures are pure, deterministic functions indexed solely by trace position. This invariant enables elimination of redundant computation across multiple diagnostic metrics through trace-level precomputation and reuse. The invariant is proven via slice-level equivalence, byte-level identity, and purity arguments, and validated with deterministic instrumentation. The full test suite remains unchanged and passes completely. The optimization reduces redundant sign computations by 75.3% within BP dynamics analysis while preserving bitwise identity, convergence behavior, and reproducibility guarantees. Scope. This document covers invariant validation and diagnostic-layer optimization only. No algorithmic changes to decoding behavior are introduced.