Home › Code › add_gumbel_noise

add_gumbel_noise

Syntologyentry name in harvested code

Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record.

add_gumbel_noise appears in the code Syntology harvested for 32 papers, as 23 distinct code bodies found in 40 places (a place is one code body under one paper). Of the 32 papers, 30 with a sample that ran: 4 with a run with no instrument failure, 26 where every run was a failure of Syntology's instrument. 18 of the code bodies carry a behaviour fingerprint.

What this page is not. Routines are grouped here by the exact string of their function or class name. Nothing asserts that two samples named add_gumbel_noise do the same thing, share code, or are comparable; the name is a string, not an identity. Behaviour outputs (what a fingerprinted sample returned on the shared battery) are not in this export and are not shown here; get_code(code_sha256="…") returns this sample's fingerprint on the shared battery, with the battery's shuffled null and lift; not shown on this site (how to connect). "Ran" means executed on a synthesized fixture, not that the code is correct or reproduces a paper.

Samples Syntology

The 23 distinct code bodies named add_gumbel_noise: 21 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 1 honoured, 0 violated, 4 with no contract checked; 16 where Syntology's instrument failed) · 2 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. One tile per status, in the site's fixed vocabulary, each code body counted once:

1ran · honoured contract
0ran · violated contract
2ran · our draft was wrong
14ran; the contract check could not run
4ran
2unverified
18fingerprinted

Licence is a property of each copy, so it is counted per place: 25 of the 40 places are pointer only (Syntology does not serve that copy's text); at 5 of those this paper's copy is under a permissive licence, but the stored copy of the same code that Syntology serves is not cleared for its text (hover the licence cell for why). This site shows no code text for any sample; every row below links to the file in its repository where the record names one.

“Ran” means the sample executed on a synthesized input; it does not mean the output is correct. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “the contract check could not run” (RAISES, DEP_MISSING or TIMEOUT; hover a row's label for which) are failures of Syntology's instrument, not of the code, and those samples did run. The ran count above is every status except unverified, the same rule as each paper page; “with no instrument failure” is honoured, violated and ran with no contract checked. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Papers

32 papers shown of 32, newest first; 40 places in the table. A paper with no recorded date is placed by the month its arXiv id encodes, shown in the Date column as YYYY-MM (from id). One row per place: a paper whose repository defines the name more than once appears more than once, and the same code body held for several papers appears once under each, with the same status. Titles and dates are the archive's archive 2025-07-28 for papers in the archive, and the graph's for 22 papers added by Syntology; 2 papers have no page here and are shown by their id only (arXiv: for an arXiv id; an id from another venue, such as ACL Anthology or AAAI, is named as that venue's id, never as an arXiv id). Status and fingerprint are Syntology's record of each code body; licence is recorded for each place. The File cell ends with the code body's code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

PaperDateFileStatus SyntologyLicence
Archer: Adaptive Reuse of Cached Hidden States for Efficient Rollback in Diffusion Language Models added by Syntology 2026-08 (from id) Hxnng/Archer/src/archer/state_distance/sampler.py 10e530678395c16a ran; the contract check could not run (RAISES) fingerprinted no licence file found · pointer only
Commit Locally, Exit Globally: Coordinating Adaptive Sampling and Early Exit in Diffusion Language Models added by Syntology 2026-07 (from id) ming053l/C4-dLLM/c4/decode.py 027b7c000eacca21 ran; the contract check could not run (RAISES) fingerprinted MIT (permissive)
Sangam: Efficiently Serving Diffusion LLMs with the AR Stack added by Syntology 2026-07 (from id) UT-InfraAI/sangam/src/sangam/sampler.py c551a291fca9e38d unverified Apache-2.0 (permissive)
Learning from the Self-future: On-policy Self-distillation for dLLMs added by Syntology 2026-06 (from id) xingzhejun/d-OPSD/d-opsd/utils.py 746963c6a9e9b2b0 ran; the contract check could not run (RAISES) no licence file found · pointer only
Learning from the Self-future: On-policy Self-distillation for dLLMs added by Syntology 2026-06 (from id) xingzhejun/d-OPSD/eval/generate.py f3233aab1c6c6543 ran; the contract check could not run (RAISES) fingerprinted no licence file found · pointer only
Prefilling-dLLM: Predictive Prefilling for Long-Context Inference in Diffusion Language Models added by Syntology 2026-06 (from id) menik1126/Prefilling-dLLM/prefilling_dllm_eval/llada_native_model.py 027b7c000eacca21 ran; the contract check could not run (RAISES) fingerprinted MIT (permissive)
SAID: Accelerating Diffusion-Based Language Models via Scaffold-Aware Iterative Decoding added by Syntology 2026-06 (from id) TH-AI-Lab-PKU/SAID/SAID-block/generate_said.py 0f2bf418118186bc ran; the contract check could not run (RAISES) fingerprinted no licence file found · pointer only
FASTKERNELS: Benchmarking GPU Kernel Generation in Production added by Syntology 2026-05 (from id) Snowflake-AI-Research/fastkernels/fastkernels/infra/dllm_engine.py e5d55fd1b07cbbf3 ran; the contract check could not run (RAISES) fingerprinted Apache-2.0 (permissive)
Roll Out and Roll Back: Diffusion LLMs are Their Own Efficiency Teachers added by Syntology 2026-05 (from id) Feng-Hong/WINO-DLLM/LLaDA/decoding.py a5334721ab55f48f ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted no licence file found · pointer only
TAD: Temporal-Aware Trajectory Self-Distillation for Fast and Accurate Diffusion LLM added by Syntology 2026-05 (from id) BHmingyang/TAD/eval/eval_llada.py 027b7c000eacca21 ran; the contract check could not run (RAISES) fingerprinted no licence file found · pointer only
Break the Block: Dynamic-size Reasoning Blocks for Diffusion Large Language Models via Monotonic Entropy Descent with Reinforcement Learning added by Syntology 2026-05 (from id) YanJiangJerry/Block-R1/rl/trainers/dynamic_generate.py 746963c6a9e9b2b0 ran; the contract check could not run (RAISES) Apache-2.0 (permissive)
R 2 -dLLM: Accelerating Diffusion Large Language Models via Spatio-Temporal Redundancy Reduction added by Syntology 2026-04 (from id) GATECH-EIC/R2-dLLM/llada/generate.py a5334721ab55f48f ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted no licence file found · pointer only
DMax: Aggressive Parallel Decoding for dLLMs added by Syntology 2026-04 (from id) czg1225/DMax/dInfer/python/dinfer/decoding/parallel_strategy.py 990cdbb7b7b9c54c ran · our draft was wrong fingerprinted Apache-2.0 (permissive)
MetaState: Persistent Working Memory Enhances Reasoning in Discrete Diffusion Language Models added by Syntology 2026-03 (from id) Les1a/MetaState/dllm/core/samplers/utils.py c7fd61b6c723375b ran; the contract check could not run (RAISES) fingerprinted Apache-2.0 (permissive)
Diffusion-State Policy Optimization for Masked Diffusion Language Models added by Syntology 2026-02 (from id) dllm-reasoning/d1/eval/generate.py f3233aab1c6c6543 ran; the contract check could not run (RAISES) fingerprinted Apache-2.0 (permissive) here; the stored copy is pointer only
Step-Wise Refusal Dynamics in Autoregressive and Diffusion Language Models added by Syntology 1 Feb 2026 niez233/DiffuGuard/utility/generate_function.py 1495eea6ced11dfa ran; the contract check could not run (RAISES) fingerprinted no licence file found · pointer only
Step-Wise Refusal Dynamics in Autoregressive and Diffusion Language Models added by Syntology 1 Feb 2026 shuita2333/PAD-codes/LLaDA-PAD/generate-PAD.py 72a32148829c38b2 ran; the contract check could not run (RAISES) fingerprinted no licence file found · pointer only
Step-Wise Refusal Dynamics in Autoregressive and Diffusion Language Models added by Syntology 1 Feb 2026 shuita2333/PAD-codes/MMaDA-PAD/models/modeling_mmada.py 615ec9b280bef7c6 ran; the contract check could not run (RAISES) fingerprinted no licence file found · pointer only
ρ-EOS: Training-free Bidirectional Variable-Length Control for Masked Diffusion LLMs added by Syntology 2026-01 (from id) yjyddq/rho-EOS/models/LLaDA_rho_EOS.py 2eceb68d6f6d3d37 ran fingerprinted Apache-2.0 (permissive)
ρ-EOS: Training-free Bidirectional Variable-Length Control for Masked Diffusion LLMs added by Syntology 2026-01 (from id) yjyddq/rho-EOS/models/LLaDA.py 3d179b42fcc82ef2 unverified Apache-2.0 (permissive)
ETS: Energy-Guided Test-Time Scaling for Training-Free RL Alignment added by Syntology 2026-01 (from id) sheriyuo/ETS/llada/ets_is.py b95cadfa99b9514d ran; the contract check could not run (RAISES) fingerprinted no licence file found · pointer only
Revealing the Attention Floating Mechanism in Masked Diffusion Models added by Syntology 2026-01 (from id) NEUIR/Attention-Floating/src/attention_extraction/Llada.py 2340e6453f04571f ran; the contract check could not run (RAISES) fingerprinted no licence file found · pointer only
CD 4 LM: Consistency Distillation and aDaptive Decoding for Diffusion Language Models added by Syntology 2026-01 (from id) yihao-liang/CDLM/evaluation/dllm_eval/models/LLaDA.py 3d179b42fcc82ef2 unverified Apache-2.0 (permissive)
KLASS: KL-Guided Fast Inference in Masked Diffusion Models added by Syntology 2025-11 (from id) identical code first harvested elsewhere a5334721ab55f48f ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted licence of this copy not recorded
Beyond Static Cutoffs: One-Shot Dynamic Thresholding for Diffusion Language Models added by Syntology 2025-11 (from id) jackshen-1215/osdt/osdt/llada_generate_osdt.py cea29bbf70d23ead ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted no licence file found · pointer only
Fast and Fluent Diffusion Language Models via Convolutional Decoding and Rejective Fine-tuning added by Syntology 2025-09 (from id) ybseo-ac/Conv/generate_yb.py a5334721ab55f48f ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted no licence file found · pointer only
Fast and Fluent Diffusion Language Models via Convolutional Decoding and Rejective Fine-tuning added by Syntology 2025-09 (from id) ybseo-ac/Conv/gen1_2_answer_generation_llada.py 529aefed4b15eb2b ran fingerprinted no licence file found · pointer only
The Devil behind the mask: An emergent safety vulnerability of Diffusion LLMs 15 Jul 2025 zichenwen1/dija/MMaDA/generate.py a5334721ab55f48f ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted Apache-2.0 (permissive) here; the stored copy is pointer only
The Devil behind the mask: An emergent safety vulnerability of Diffusion LLMs 15 Jul 2025 zichenwen1/dija/MMaDA/models/modeling_mmada.py 615ec9b280bef7c6 ran; the contract check could not run (RAISES) fingerprinted Apache-2.0 (permissive) here; the stored copy is pointer only
arXiv:2506.15735 2025-06 (from id) lasr-eliciting-contexts/ContextBench/src/contextbench/llada/generate.py a5334721ab55f48f ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted no licence file found · pointer only
dLLM-Cache: Accelerating Diffusion Large Language Models with Adaptive Caching 17 May 2025 maomaocun/dLLM-cache/demo_MMada_cache.py 008ab2cfccca5ac2 ran fingerprinted Apache-2.0 (permissive)
dLLM-Cache: Accelerating Diffusion Large Language Models with Adaptive Caching 17 May 2025 maomaocun/dLLM-cache/demo_MMada_mmu_cache.py 22cb8c51f0a9038f ran fingerprinted Apache-2.0 (permissive)
MMaDA: Multimodal Large Diffusion Language Models 21 May 2025 Gen-Verse/MMaDA/generate.py a5334721ab55f48f ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted MIT (permissive) here; the stored copy is pointer only
MMaDA: Multimodal Large Diffusion Language Models 21 May 2025 Gen-Verse/MMaDA/models/modeling_mmada.py 615ec9b280bef7c6 ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) here; the stored copy is pointer only
Large Language Diffusion Models 14 Feb 2025 ml-gsai/llada/generate.py a5334721ab55f48f ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted no licence file found · pointer only
ENAT: Rethinking Spatial-temporal Interactions in Token-based Image Synthesis 11 Nov 2024 leaplabthu/enat/libs/nat_misc.py 9ef9985f5609c318 ran · our draft was wrong no licence file found · pointer only
Revisiting Non-Autoregressive Transformers for Efficient Image Synthesis 8 Jun 2024 leaplabthu/improvednat/libs/nat_misc.py 9ef9985f5609c318 ran · our draft was wrong MIT (permissive)
REMARK-LLM: A Robust and Efficient Watermarking Framework for Generative Large Language Models 18 Oct 2023 ruisizhang123/REMARK-LLM/inference.py 2d1739ff0a683922 ran · honoured contract no licence file found · pointer only
Muse: Text-To-Image Generation via Masked Generative Transformers 2 Jan 2023 baaivision/muse-pytorch/libs/muse.py 9ef9985f5609c318 ran · our draft was wrong MIT (permissive)
CVF open access id Hertz_Style_Aligned_Image_Generation_via_Shared_Attention_CVPR_2024_paper baaivision/MUSE-Pytorch/libs/muse.py 9ef9985f5609c318 ran · our draft was wrong MIT (permissive)

This site shows no code text; each File cell links to the file on GitHub at the repository's current default branch, which may have changed since the harvest. "Pointer only" means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is under a permissive licence, but the stored copy of the same code that Syntology serves is not cleared for its text. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence cell for the reason. Per-sample records for a paper are on its paper page under "Code Syntology ran".

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections