Warren Johnson
6 of 8 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (6 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 6 have no contract verdict; 0 failures of our instrument; 2 no recorded run), measured 2026-11-22; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.
"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2602.15843.json · badge: /api/ran/2602.15843.svg
Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.
| Repository | Role | Ran |
|---|---|---|
| micoverde/taac-llm-compression | community Papers with Code archive |
6 ran · 2 no recorded run |
| Function | Status | Where it lives |
|---|---|---|
| CompressionEngine | Ran ran on a synthesized input · constructed an object; no result was computed |
micoverde/taac-llm-compression/taac_algorithm.py pointer only (licence: NONE) · get_code("7adb3cae4aee7460") How it ranThe body of this sample is withheld: GitHub detected no licence in this paper's repository, micoverde/taac-llm-compression when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded. |
| CompressionResult | Ran ran on a synthesized input · constructed an object; no result was computed |
micoverde/taac-llm-compression/taac_algorithm.py pointer only (licence: NONE) · get_code("bf9d7a1bfe3a269c") How it ranThe body of this sample is withheld: GitHub detected no licence in this paper's repository, micoverde/taac-llm-compression when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded. |
| InformationDensityEstimator | Ran ran on a synthesized input · constructed an object; no result was computed |
micoverde/taac-llm-compression/taac_algorithm.py pointer only (licence: NONE) · get_code("e50f758640eba88a") How it ranThe body of this sample is withheld: GitHub detected no licence in this paper's repository, micoverde/taac-llm-compression when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded. |
| QualityPredictor | Ran ran on a synthesized input · constructed an object; no result was computed |
micoverde/taac-llm-compression/taac_algorithm.py pointer only (licence: NONE) · get_code("cadacdca25e9773e") How it ranThe body of this sample is withheld: GitHub detected no licence in this paper's repository, micoverde/taac-llm-compression when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded. |
| TAACConfig | Ran ran on a synthesized input · constructed an object; no result was computed |
micoverde/taac-llm-compression/taac_algorithm.py pointer only (licence: NONE) · get_code("5c6e03bbefa632f5") How it ranThe body of this sample is withheld: GitHub detected no licence in this paper's repository, micoverde/taac-llm-compression when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded. |
| TaskClassifier | Ran ran on a synthesized input · constructed an object; no result was computed |
micoverde/taac-llm-compression/taac_algorithm.py pointer only (licence: NONE) · get_code("f29a0b9c383a67de") How it ranThe body of this sample is withheld: GitHub detected no licence in this paper's repository, micoverde/taac-llm-compression when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded. |
| TAAC | Not yet run unverified (harvested, no recorded run) |
micoverde/taac-llm-compression/taac_algorithm.py pointer only (licence: NONE) · get_code("3738975a0f3acd9e") How it ranThe body of this sample is withheld: GitHub detected no licence in this paper's repository, micoverde/taac-llm-compression when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded. |
| TaskType | Not yet run unverified (harvested, no recorded run) |
micoverde/taac-llm-compression/taac_algorithm.py pointer only (licence: NONE) · get_code("b751c25eb91073ad") How it ranThe body of this sample is withheld: GitHub detected no licence in this paper's repository, micoverde/taac-llm-compression when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded. |
Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 8 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).
Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.
Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
In "Compress or Route?" (Johnson, 2026), we found that code generation and chain-of-thought reasoning respond differently to prompt compression: code tolerates aggressive compression (r ≥ 0.6) while reasoning degrades gradually. That initial study, however, was limited to a single code benchmark (HumanEval, 164 problems), left the proposed "perplexity paradox" mechanism unvalidated, and provided no adaptive algorithm. This paper addresses all three gaps. First, we validate the task-dependent compression hypothesis across six code benchmarks (HumanEval, MBPP, HumanEval+, MultiPL-E in Python/JavaScript/Java) and four reasoning benchmarks (GSM8K, MATH, ARC-Challenge, MMLU-STEM), demonstrating that the compression threshold (r ≥ 0.6) generalizes across programming languages and problem difficulties. Second, we conduct the first per-token perplexity analysis of compression decisions (n = 723 tokens), revealing a "perplexity paradox": code syntax tokens (which appear unusual to language models) are preserved, while numerical values in math problems (which follow predictable syntactic patterns) are pruned despite being task-critical. In a controlled signature preservation experiment, we demonstrate a +34 percentage point recovery in pass rate (5.3% baseline → 39.3% with signature injection; Cohen's h = 0.890, very large effect), with NameError rates dropping from 86.1% to 6.1%. Third, we propose TAAC (Task-Aware Adaptive Compression), a quality-gated algorithm that dynamically adjusts compression based on predicted quality degradation, achieving 22% cost reduction with 96% quality preservation-outperforming fixed-ratio compression by 7%. Our MBPP validation (n = 1,800 trials across 6 compression ratios) confirms compression tolerance varies systematically with ratio: 3.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2602.15843")
get_code_for_paper("2602.15843")
have("2602.15843")
The run record, dated, one paper per request, free:
curl https://syntology.ai/api/ran/2602.15843.json
A badge for a README (the split and the date, never a ratio):
[](https://syntology.ai/paper/2602.15843)
Connect an agent — have() is free.