SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2402.11809 · 2024

Generation Meets Verification: Accelerating Large Language Model Inference with Smart Parallel Auto-Correct Decoding

arXiv · PDF · Open in the Atlas

Code that ran

9 of 11 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 1 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 8 failures of our instrument; 2 no recorded run); 1 of them was also run on the host, not in the jail, by the peer-agreement lane that set its level, measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; 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/2402.11809.json · badge: /api/ran/2402.11809.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
cteant/space canonical (classifier label)
classifier basis: census
9 ran · 2 no recorded run
FunctionStatusWhere it lives
repeat_kv Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
cteant/space/src/models/modeling_llama.py
code served (permissive licence) · get_code("30d7eec482ebf6b1")
How it ran

Ran in Syntology's jail on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned Tensor, ndim 4, dtype family float, shape (2, 6, 5, 4) as a later run recorded it. The image it ran in was not recorded. The comparison behind this sample's level is a different run: the peer-agreement lane ran it again on the host, not in the jail (report dated 2026-09-04T23:09:20Z). If your copy returns the same type, ndim and shape 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

torch.manual_seed(0)

# Create a tensor with shape (batch, num_key_value_heads, seqlen, head_dim)
# Using distinct sizes: batch=2, num_key_value_heads=3, seqlen=5, head_dim=4
hidden_states = torch.randn(2, 3, 5, 4)

# n_rep is the repetition factor
n_rep = 2

ARGS = {
    "hidden_states": hidden_states,
    "n_rep": n_rep,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 4, dtype family float, shape (2, 6, 5, 4) — shape (2, 6, 5, 4) as a later run recorded it
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
apply_rotary_pos_emb Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
cteant/space/src/models/modeling_chatglm.py
code served (permissive licence) · get_code("5cfeaee89a75c5ff")
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) and returned Tensor, ndim 4, dtype family float, shape (3, 2, 4, 8) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape 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

torch.manual_seed(0)

# Analyze the function:
# - x: torch.Tensor with shape (sq, b, np, hn)
#   - sq: sequence length
#   - b: batch size
#   - np: number of heads
#   - hn: head dimension
# - rope_cache: torch.Tensor with shape (seq_len, ..., rot_dim//2, 2)
#   - Used as rope_cache[:sq] and rope_cache.view(sq, -1, 1, xshaped.size(3), 2)
#   - rot_dim = rope_cache.shape[-2] * 2

# Let's construct valid tensors:
# sq = 3 (sequence length)
# b = 2 (batch size)
# np = 4 (number of heads)
# hn = 8 (head dimension)
# rot_dim = 6 (so rope_cache.shape[-2] = 3)

sq = 3
b = 2
np = 4
hn = 8
rot_dim = 6

# x shape: (sq, b, np, hn) = (3, 2, 4, 8)
x = torch.randn(sq, b, np, hn)

# rope_cache shape needs to support:
# - rope_cache[:sq] indexing
# - rope_cache.view(sq, -1, 1, xshaped.size(3), 2)
# - rope_cache.shape[-2] * 2 = rot_dim
# So rope_cache.shape[-2] = 3
# xshaped.size(3) = rot_dim // 2 = 3
# After reshape: xshaped = (sq, -1, np, rot_dim//2, 2) = (3, -1, 4, 3, 2)
# So the middle dimension is b = 2
# rope_cache needs shape (seq_len, b, 1, 3, 2) or compatible
# Let's make it (seq_len, b, 3, 2) and it will view to (sq, b, 1, 3, 2)

rope_cache = torch.randn(sq + 1, b, rot_dim // 2, 2)

ARGS = {
    "x": x,
    "rope_cache": rope_cache,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 4, dtype family float, shape (3, 2, 4, 8) — shape (3, 2, 4, 8) as a later run recorded it
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
build_alibi_tensor Ran
ran; our contract draft was wrong, not the code
cteant/space/src/models/modeling_falcon.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from bkhmsi/llm-localization, is pointer only (licence: NONE) · get_code("b4acc4f0962c97ca")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from bkhmsi/llm-localization, and GitHub detected no licence in that repository 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.

dequantize_cache_torch Ran
ran, honoured the contract we drafted
cteant/space/src/models/modeling_qwen.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from qwenlm/qwen-audio, is pointer only (licence: NOASSERTION) · get_code("d9db9f86b30f9ac0")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from qwenlm/qwen-audio, and GitHub could not classify the licence of that repository (NOASSERTION). 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.

dropout_add Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
cteant/space/src/models/modeling_falcon.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from damo-nlp-sg/multilingual_analysis, is pointer only (licence: NONE) · get_code("546c22d62993a619")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from damo-nlp-sg/multilingual_analysis, and GitHub detected no licence in that repository 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.

quantize_cache_v Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
cteant/space/src/models/modeling_qwen.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from qwenlm/qwen-audio, is pointer only (licence: NOASSERTION) · get_code("1fec7a76f0d171a6")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from qwenlm/qwen-audio, and GitHub could not classify the licence of that repository (NOASSERTION). 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.

rotate_half Ran
ran; our contract draft was wrong, not the code
cteant/space/src/models/modeling_internlm.py
code served (permissive licence) · get_code("b99eea6376d1e212")
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) and returned Tensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape 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

torch.manual_seed(0)

# The function rotate_half takes a single parameter 'x' which is a tensor
# It accesses x.shape[-1] to get the last dimension
# It uses indexing with [..., :] slicing notation
# It calls torch.cat() to concatenate tensors

# Create a small tensor with shape (2, 4) - batch size 2, hidden dim 4
x = torch.randn(2, 4)

ARGS = {
    "x": x
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
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
rotate_half Ran
ran; our contract draft was wrong, not the code
cteant/space/src/models/modeling_falcon.py
code served (permissive licence) · get_code("c89eed9eaa3332c2")
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 Tensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape 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

torch.manual_seed(0)

# The function rotate_half takes a tensor x and:
# 1. Splits it along the last dimension into two halves
# 2. Concatenates them in reversed order with negation on the first half

# Create a small tensor with batch size 2 and feature dimension 4
x = torch.randn(2, 4)

ARGS = {
    "x": x
}
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 Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
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
split_tensor_along_last_dim Ran
ran; our contract draft was wrong, not the code
cteant/space/src/models/modeling_chatglm.py
code served (permissive licence) · get_code("d7e378ee7899fbde")
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) and returned tuple, ndim 4, 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

torch.manual_seed(0)

# Create a small tensor with shape (2, 4, 8)
# Last dimension is 8, which is divisible by 2 and 4
tensor = torch.randn(2, 4, 8)

ARGS = {
    "tensor": tensor,
    "num_partitions": 2,
    "contiguous_split_chunks": False,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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 4, 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
apply_rotary_pos_emb Not yet run
unverified (harvested, no recorded run)
cteant/space/src/models/modeling_internlm.py
code served (permissive licence) · get_code("b42659b1a3b456f9")
How it ran

no invocation recorded.

default_init Not yet run
unverified (harvested, no recorded run)
cteant/space/src/models/modeling_chatglm.py
code served (permissive licence) · get_code("237019e36b9481d9")
How it ran

no invocation recorded.

Also present in other papers' code

Of 11 samples lifted from this paper's linked repositories, 11 carry a function identity and 0 do not. Of the 11 distinct functions with an identity, 10 are also present in other papers whose code we hold (10 with at least one paper dated earlier, 4 the same month, 10 later) and 1 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
rotate_half 360 other papers
68 dated earlier, 25 the same month, 258 later, 6 same year, order not determinable, 3 undated · 30 only as an assembly brick
1901.02860 (2019-01)
Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context
2203.08913 (2022-03)
Memorizing Transformers
2302.13971 (2023-02)
LLaMA: Open and Efficient Foundation Language Models
and 357 more not named here
repeat_kv 243 other papers
33 dated earlier, 19 the same month, 187 later, 2 same year, order not determinable, 2 undated · 22 only as an assembly brick
1901.02860 (2019-01)
Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context
2203.08913 (2022-03)
Memorizing Transformers
2305.17888 (2023-05)
LLM-QAT: Data-Free Quantization Aware Training for Large Language Models
and 240 more not named here
dropout_add 8 other papers
3 dated earlier, 1 the same month, 2 later, 1 same year, order not determinable, 1 undated
2311.04219 (2023-11)
OtterHD: A High-Resolution Multi-modality Model
2311.13230 (2023-11)
Enhancing Uncertainty-Based Hallucination Detection with Stronger Focus
2024.findings-acl.347 (2024 (year only))
Mass-Editing Memory with Attention in Transformers: A cross-lingual exploration of knowledge
and 5 more not named here
rotate_half 8 other papers
2 dated earlier, 1 the same month, 5 later · 1 only as an assembly brick
2310.16834 (2023-10)
Discrete Diffusion Modeling by Estimating the Ratios of the Data Distribution
2401.12522 (2024-01)
BiTA: Bi-Directional Tuning for Lossless Acceleration in Large Language Models
2402.07754 (2024-02)
Diffusion of Thought: Chain-of-Thought Reasoning in Diffusion Language Models
and 5 more not named here
split_tensor_along_last_dim 8 other papers
3 dated earlier, 5 later
2307.13528 (2023-07)
FacTool: Factuality Detection in Generative AI -- A Tool Augmented Framework for Multi-Task and Multi-Domain Scenarios
2311.18445 (2023-11)
VTimeLLM: Empower LLM to Grasp Video Moments
2401.18058 (2024-01)
LongAlign: A Recipe for Long Context Alignment of Large Language Models
and 5 more not named here
build_alibi_tensor 6 other papers
3 dated earlier, 2 later, 1 same year, order not determinable
2311.04219 (2023-11)
OtterHD: A High-Resolution Multi-modality Model
2311.13230 (2023-11)
Enhancing Uncertainty-Based Hallucination Detection with Stronger Focus
2024.findings-acl.347 (2024 (year only))
Mass-Editing Memory with Attention in Transformers: A cross-lingual exploration of knowledge
and 3 more not named here
default_init 6 other papers
3 dated earlier, 3 later
2307.13528 (2023-07)
FacTool: Factuality Detection in Generative AI -- A Tool Augmented Framework for Multi-Task and Multi-Domain Scenarios
2311.18445 (2023-11)
VTimeLLM: Empower LLM to Grasp Video Moments
2401.18058 (2024-01)
LongAlign: A Recipe for Long Context Alignment of Large Language Models
and 3 more not named here
apply_rotary_pos_emb 4 other papers
3 dated earlier, 1 later
2307.13528 (2023-07)
FacTool: Factuality Detection in Generative AI -- A Tool Augmented Framework for Multi-Task and Multi-Domain Scenarios
2311.18445 (2023-11)
VTimeLLM: Empower LLM to Grasp Video Moments
2401.18058 (2024-01)
LongAlign: A Recipe for Long Context Alignment of Large Language Models
and 1 more not named here
dequantize_cache_torch 3 other papers
1 dated earlier, 2 later
2311.07919 (2023-11)
Qwen-Audio: Advancing Universal Audio Understanding via Unified Large-Scale Audio-Language Models
2406.03872 (2024-06)
BLSP-Emo: Towards Empathetic Large Speech-Language Models
2410.03007 (2024-10)
FastAdaSP: Multitask-Adapted Efficient Inference for Large Speech Language Model
quantize_cache_v 3 other papers
1 dated earlier, 2 later
2311.07919 (2023-11)
Qwen-Audio: Advancing Universal Audio Understanding via Unified Large-Scale Audio-Language Models
2406.03872 (2024-06)
BLSP-Emo: Towards Empathetic Large Speech-Language Models
2410.03007 (2024-10)
FastAdaSP: Multitask-Adapted Efficient Inference for Large Speech Language Model

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

This research aims to accelerate the inference speed of large language models (LLMs) with billions of parameters. We propose \textbf{S}mart \textbf{P}arallel \textbf{A}uto-\textbf{C}orrect d\textbf{E}coding (SPACE), an innovative approach designed for achieving lossless acceleration of LLMs. By integrating semi-autoregressive inference and speculative decoding capabilities, SPACE uniquely enables autoregressive LLMs to parallelize token generation and verification. This is realized through a specialized semi-autoregressive supervised fine-tuning process that equips existing LLMs with the ability to simultaneously predict multiple tokens. Additionally, an auto-correct decoding algorithm facilitates the simultaneous generation and verification of token sequences within a single model invocation. Through extensive experiments on a range of LLMs, SPACE has demonstrated inference speedup ranging from 2.7x-4.0x on HumanEval-X while maintaining output quality.

For agents

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

get_harvested_code_for_paper("2402.11809")
get_code_for_paper("2402.11809")
have("2402.11809")

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

curl https://syntology.ai/api/ran/2402.11809.json

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

[![Syntology run record](https://syntology.ai/api/ran/2402.11809.svg)](https://syntology.ai/paper/2402.11809)

Connect an agent — have() is free.