SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2509.26432 · ICLR · 2025

AdaBlock-dLLM: SEMANTIC-AWARE DIFFUSION LLM INFERENCE VIA ADAPTIVE BLOCK SIZE

Mark Chen, Guanxi Lu, Hongxiang Fan, Yuto Karashima, Zhican Wang, Daichi Fujiki

arXiv · PDF · Open in the Atlas

Code that ran

3 of 4 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (2 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 2 have no contract verdict; 1 failure of our instrument; 1 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/2509.26432.json · badge: /api/ran/2509.26432.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.

RepositoryRoleRan
lgxi24/AdaBlock-dLLM community
Papers with Code archive
3 ran · 1 no recorded run
FunctionStatusWhere it lives
DreamGenerationConfig Ran
ran on a synthesized input · constructed an object; no result was computed
lgxi24/AdaBlock-dLLM/dream/model/generation_utils_adablock.py
code served (permissive licence) · get_code("f484db2354d52ac3")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, DreamGenerationConfig: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
from transformers.generation.configuration_utils import GenerationConfig
from transformers import __version__

torch.manual_seed(0)

ARGS = {
    'temperature': 0.9,
    'top_p': 0.95,
    'top_k': 50,
    'max_length': 128,
    'max_new_tokens': 50,
    'eps': 0.001,
    'steps': 512,
    'alg': 'origin',
    'alg_temp': 0.8,
    'num_return_sequences': 2,
    'return_dict_in_generate': True,
    'output_history': False,
    'mask_token_id': 103,
    'pad_token_id': 0,
    'bos_token_id': 101,
    'eos_token_id': 102,
    'generation_kwargs': {'num_beams': 1},
    '_from_model_config': False,
    '_commit_hash': 'abc123def456',
    'transformers_version': __version__,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, transformers — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DreamGenerationConfig — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a DreamGenerationConfig output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
DreamModelOutput Ran
ran on a synthesized input · constructed an object; no result was computed
lgxi24/AdaBlock-dLLM/dream/model/generation_utils_adablock.py
code served (permissive licence) · get_code("f375df03b30b330d")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, DreamModelOutput: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
from transformers.utils import ModelOutput

torch.manual_seed(0)

ARGS = {
    "sequences": torch.LongTensor([[1, 2, 3, 4], [5, 6, 7, 8]]),
    "history": (torch.FloatTensor([[0.1, 0.2, 0.3, 0.4], [0.5, 0.6, 0.7, 0.8]]),),
    "nfe_history": [1, 2, 3, 4],
    "block_history": [{"block": 0}, {"block": 1}],
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, transformers — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DreamModelOutput — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a DreamModelOutput output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
sample_tokens Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
lgxi24/AdaBlock-dLLM/dream/model/generation_utils_adablock.py
code served (permissive licence) · get_code("bf0792b4bae5f006")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple, ndim 2, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
import torch.nn.functional as F

torch.manual_seed(0)

# Create a small logits tensor
# Shape: (batch_size=2, vocab_size=4)
logits = torch.randn(2, 4)

ARGS = {
    "logits": logits,
    "temperature": 0.7,
    "top_p": 0.9,
    "top_k": 3,
    "margin_confidence": False,
    "neg_entropy": False,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 2, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
DreamGenerationMixin Not yet run
unverified (harvested, no recorded run)
lgxi24/AdaBlock-dLLM/dream/model/generation_utils_adablock.py
code served (permissive licence) · get_code("d04f7ce39ab3418f")
How it ran

no invocation recorded.

Also present in other papers' code

Of 4 samples lifted from this paper's linked repositories, 4 carry a function identity and 0 do not. Of the 4 distinct functions with an identity, 1 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 1 later) and 3 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.

FunctionAlso present inEarliest-datedOthers named
sample_tokens
brick of this paper's assembly: sampler
5 other papers
5 later · 1 only as an assembly brick
2602.01326 (2026-02)
DreamOn: Diffusion Language Models For Code Infilling Beyond Fixed-size Canvas
2602.06953 (2026-02)
DAWN: Dependency-Aware Fast Inference for Diffusion LLMs
2604.15750 (2026-04)
DepCap: Adaptive Block-Wise Parallel Decoding for Efficient Diffusion LM Inference
and 2 more not named here

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.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Diffusion-based large language models (dLLMs) are gaining attention for their inherent capacity for parallel decoding, offering a compelling alternative to autoregressive LLMs. Among various decoding strategies, block-wise semiautoregressive (semi-AR) approaches are widely adopted due to their support for KV caching and their favorable accuracy-speed trade-off. However, this paper identifies two fundamental limitations in the conventional semi-AR decoding approach that applies a fixed block size: i) late decoding overhead, where the unmasking of high-confidence tokens outside the current block is unnecessarily delayed, and ii) premature decoding error, where low-confidence tokens inside the current block are committed too early, leading to incorrect tokens. This paper presents the first systematic investigation challenging the fixed block size setting in semi-AR decoding. Through a statistical analysis of confidence dynamics during the denoising process, we identify a volatility band (VB) region during dLLM decoding, which encodes local semantic structure and can be used to guide adaptive block sizing. Leveraging these insights, we introduce AdaBlock-dLLM, a training-free, plug-and-play scheduler that adaptively aligns block boundaries with semantic steps by adjusting block size during runtime. Extensive experiments across diverse benchmarks show that AdaBlock-dLLM achieves up to 5.3% accuracy improvement under the same throughput budget. Beyond inferencetime optimization, we hope our semantics-aware adaptive scheduling approach and confidence-based analysis will inspire future training strategies for dLLMs. Our code is available at https://github.com/lgxi24/AdaBlock-dLLM.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2509.26432")
get_code_for_paper("2509.26432")
have("2509.26432")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2509.26432.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2509.26432.svg)](https://syntology.ai/paper/2509.26432)

Connect an agent — have() is free.