SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2406.15765 · 2024

Unveiling and Harnessing Hidden Attention Sinks: Enhancing Large Language Models without Training through Attention Calibration

arXiv · PDF · Open in the Atlas

Code that ran

5 of 13 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; 0 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 5 failures of our instrument; 8 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/2406.15765.json · badge: /api/ran/2406.15765.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
gatech-eic/act canonical (classifier label)
classifier basis: census
5 ran · 8 no recorded run
FunctionStatusWhere it lives
repeat_kv Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
gatech-eic/act/models/modelling_llama.py
pointer only (licence: NONE) · get_code("30d7eec482ebf6b1")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, gatech-eic/act 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.

apply_rotary_pos_emb Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
gatech-eic/act/models/modelling_llama.py
pointer only (licence: NONE) · get_code("d61c483a3c2b3156")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, gatech-eic/act 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.

atten_process_cal Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
gatech-eic/act/models/llama_modelling_aug.py
pointer only (licence: NONE) · get_code("22be0f0263f68218")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, gatech-eic/act 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.

form_choices Ran
ran; our contract draft was wrong, not the code
gatech-eic/act/utils/data_utils.py
pointer only (licence: NONE) · get_code("fb732786766c9170")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, gatech-eic/act 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.

rotate_half Ran
ran; our contract draft was wrong, not the code
gatech-eic/act/models/modelling_llama.py
pointer only (licence: NONE) · get_code("b99eea6376d1e212")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, gatech-eic/act 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.

atten_aug_forward_eval_llama Not yet run
unverified (harvested, no recorded run)
gatech-eic/act/models/llama_modelling_aug.py
pointer only (licence: NONE) · get_code("81cfc398d4db8c00")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, gatech-eic/act 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.

atten_aug_forward_eval_mistral Not yet run
unverified (harvested, no recorded run)
gatech-eic/act/models/mistral_modelling_aug.py
pointer only (licence: NONE) · get_code("8ec16c9a57666186")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, gatech-eic/act 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.

atten_aug_forward_llama Not yet run
unverified (harvested, no recorded run)
gatech-eic/act/mt-bench/gen_model_answer.py
pointer only (licence: NONE) · get_code("4491e28ac0c62c55")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, gatech-eic/act 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.

atten_process Not yet run
unverified (harvested, no recorded run)
gatech-eic/act/mt-bench/gen_model_answer.py
pointer only (licence: NONE) · get_code("9766c8d6378000f2")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, gatech-eic/act 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.

atten_process_eval Not yet run
unverified (harvested, no recorded run)
gatech-eic/act/models/llama_modelling_aug.py
pointer only (licence: NONE) · get_code("d20d6b485da65eef")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, gatech-eic/act 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.

atten_process_eval Not yet run
unverified (harvested, no recorded run)
gatech-eic/act/models/mistral_modelling_aug.py
pointer only (licence: NONE) · get_code("936d7a36c6e12df6")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, gatech-eic/act 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.

get_classification_dataset Not yet run
unverified (harvested, no recorded run)
gatech-eic/act/utils/data_utils.py
pointer only (licence: NONE) · get_code("5d8307adf1e23d57")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, gatech-eic/act 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.

get_multiple_choice_dataset Not yet run
unverified (harvested, no recorded run)
gatech-eic/act/utils/data_utils.py
pointer only (licence: NONE) · get_code("1bea206b322ec46d")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, gatech-eic/act 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.

Also present in other papers' code

Of 13 samples lifted from this paper's linked repositories, 13 carry a function identity and 0 do not. Of the 13 distinct functions with an identity, 3 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 3 the same month, 3 later) and 10 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
135 dated earlier, 21 the same month, 195 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
79 dated earlier, 16 the same month, 144 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
apply_rotary_pos_emb 75 other papers
31 dated earlier, 4 the same month, 40 later · 2 only as an assembly brick
2309.11499 (2023-09)
DreamLLM: Synergistic Multimodal Comprehension and Creation
2312.02051 (2023-12)
TimeChat: A Time-sensitive Multimodal Large Language Model for Long Video Understanding
2312.03414 (2023-12)
Compressed Context Memory For Online Language Model Interaction
and 72 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

Attention is a fundamental component behind the remarkable achievements of large language models (LLMs). However, our current understanding of the attention mechanism, especially regarding how attention distributions are established, remains limited. Inspired by recent studies that explore the presence of attention sink in the initial token, which receives disproportionately large attention scores despite their lack of semantic importance, this work delves deeper into this phenomenon. We aim to provide a more profound understanding of the existence of attention sinks within LLMs and to uncover ways to enhance the achievable accuracy of LLMs by directly optimizing the attention distributions, without the need for weight finetuning. Specifically, this work begins with comprehensive visualizations of the attention distributions in LLMs during inference across various inputs and tasks. Based on these visualizations, to the best of our knowledge, we are the first to discover that (1) attention sinks occur not only at the start of sequences but also within later tokens of the input, and (2) not all attention sinks have a positive impact on the achievable accuracy of LLMs. Building upon our findings, we propose a training-free Attention Calibration Technique (ACT) that automatically optimizes the attention distributions on the fly during inference in an input-adaptive manner. Extensive experiments validate that ACT consistently enhances the accuracy of various LLMs across different applications. Specifically, ACT achieves an average improvement of up to 7.30% in accuracy across different datasets when applied to Llama-30B. Our code is available at https://github.com/GATECH-EIC/ACT.

For agents

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

get_harvested_code_for_paper("2406.15765")
get_code_for_paper("2406.15765")
have("2406.15765")

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

curl https://syntology.ai/api/ran/2406.15765.json

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

[![Syntology run record](https://syntology.ai/api/ran/2406.15765.svg)](https://syntology.ai/paper/2406.15765)

Connect an agent — have() is free.