Papers › Disentangling and Integrating Relational and Sensory Information in Transformer Architectures

Disentangling and Integrating Relational and Sensory Information in Transformer Architectures

26 May 2024arXiv:2405.16727archive 2025-07-28

Awni Altabaa, John Lafferty

Relational reasoning is a central component of generally intelligent systems, enabling robust and data-efficient inductive generalization. Recent empirical evidence shows that many existing neural architectures, including Transformers, struggle with tasks requiring relational reasoning. In this work, we distinguish between two types of information: sensory information about the properties of individual objects, and relational information about the relationships between objects. While neural attention provides a powerful mechanism for controlling the flow of sensory information between objects, the Transformer lacks an explicit computational mechanism for routing and processing relational information. To address this limitation, we propose an architectural extension of the Transformer framework that we call the Dual Attention Transformer (DAT), featuring two distinct attention mechanisms: sensory attention for directing the flow of sensory information, and a novel relational attention mechanism for directing the flow of relational information. We empirically evaluate DAT on a diverse set of tasks ranging from synthetic relational benchmarks to complex real-world tasks such as language modeling and visual processing. Our results demonstrate that integrating explicit relational computational mechanisms into the Transformer architecture leads to significant performance gains in terms of data efficiency and parameter efficiency.

PaperPDFCodeCode Syntology ranMethod as code

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 21 code samples harvested from 2 repositories linked to this paper: 14 ran (of which 9 constructed an object rather than computing a result; 10 with no instrument failure: 0 honoured, 1 violated, 9 with no contract checked; 4 where Syntology's instrument failed) · 7 unverified. Where Syntology's instrument failed: 3 where our contract draft was wrong; 1 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 6 samples from 1 repository, 2 ran (1 with no instrument failure; 1 where Syntology's instrument failed); community (archive-listed): 15 samples from 1 repository, 12 ran (9 with no instrument failure; 3 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written on 2026-09-22 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

awni00/dual-attention officialmentioned in papermentioned on GitHubpytorchSyntology: 6 samples, 2 ran (1 instrument-failed) · 4 unverified report
awni00/abstract_transformer mentioned on GitHubpytorchSyntology: 15 samples, 12 ran (3 instrument-failed) · 3 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

21 samples harvested: 14 ran (of which 9 constructed an object rather than computing a result; 10 with no instrument failure: 0 honoured, 1 violated, 9 with no contract checked; 4 where Syntology's instrument failed) · 7 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · violated contract
3ran · our draft was wrong
1ran; the contract check could not run
9ran
7unverified

Licence: 0 of the 21 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 2 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. For 6 samples that ran, the graph records a signature this page does not show: for a class the fixture built without calling, the harness called the object with inputs it made up itself, so the recorded signature is the harness's probe, not the fixture's arguments. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its 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.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at 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 cleared, but the stored copy of the same code that Syntology serves is not. 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 label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

awni00/dual-attention official repository 6 samples, 2 ran (1 instrument-failed) · 4 unverified
RelationalAttention dual_attention/dual_attention.py ran MIT (permissive) · e96ef6f7fa7fb78b · report fixture: returned RelationalAttention · level recorded 2026-09-22
apply_rotary_emb in 121 papers dual_attention/dual_attention.py ran · our draft was wrong MIT (permissive) · dd15a61ea3d0e604 · report fixture: signature xq:Tensor/3/float, xk:Tensor/3/float, freqs_cos:Tensor/2/float, freqs_sin:Tensor/2/float · returned tuple · level recorded 2026-09-22
DisentangledRelationalCrossAttention dual_attention/dual_attention.py unverified MIT (permissive) · 65050ac2515e807a · report
DualAttention dual_attention/dual_attention.py unverified MIT (permissive) · 906b56a4d10f246c · report
RelationalCrossAttention dual_attention/dual_attention.py unverified MIT (permissive) · 8f6de9b1d93efd34 · report
awni00/abstract_transformer community (archive-listed) 15 samples, 12 ran (3 instrument-failed) · 3 unverified
Attention in 344 papers
census: 13 implementations from 12 papers, 10 distinct outputs across 4 buckets, one shared input per bucket
dual_attention_transformer.py ran · metamorphic tier: deterministic MIT (permissive) · 437880805ab90b88 · report fixture: returned Attention · level recorded 2026-09-22
FeedForwardBlock in 7 papers dual_attention_transformer.py ran · metamorphic tier: deterministic MIT (permissive) · a001e1fb0893402a · report fixture: returned FeedForwardBlock · level recorded 2026-09-22
PositionRelativeSymbolRetriever dual_attention_transformer.py ran · metamorphic tier: deterministic fingerprinted MIT (permissive) · 27aa5362b0778651 · report fixture: returned PositionRelativeSymbolRetriever · level recorded 2026-09-22
PositionalSymbolRetriever dual_attention_transformer.py ran · metamorphic tier: deterministic fingerprinted MIT (permissive) · 07d2253837bc4d3f · report fixture: returned PositionalSymbolRetriever · level recorded 2026-09-22
RMSNorm in 61 papers dual_attention_transformer.py ran · metamorphic tier: deterministic MIT (permissive) · 01777e072296c0bb · report fixture: returned RMSNorm · level recorded 2026-09-22
RelationalAttention dual_attention_transformer.py ran · metamorphic tier: deterministic MIT (permissive) · f75b1ab91cff5f54 · report fixture: returned RelationalAttention · level recorded 2026-09-22
RelativePositionalEncoding dual_attention_transformer.py ran · metamorphic tier: deterministic MIT (permissive) · 41fd4f0515e6826a · report fixture: returned RelativePositionalEncoding · level recorded 2026-09-22
SymbolicAttention dual_attention_transformer.py ran · metamorphic tier: deterministic MIT (permissive) · 32699e73b61b32e9 · report fixture: returned SymbolicAttention · level recorded 2026-09-22
apply_rotary_emb in 121 papers dual_attention_transformer.py ran; the contract check could not run (RAISES) MIT (permissive) · 6dc1f541b73c3081 · report fixture: signature xq:Tensor/3/float, xk:Tensor/3/float, freqs_cos:Tensor/2/float, freqs_sin:Tensor/2/float · returned tuple · level recorded 2026-09-22
compute_causal_mask dual_attention_transformer.py ran · violated contract fingerprinted MIT (permissive) · a73cb1ba115d3af9 · report fixture: signature size:int/0/int, device:NoneType/0/? · returned Tensor · level recorded 2026-09-22
create_norm in 8 papers dual_attention_transformer.py ran · our draft was wrong MIT (permissive) · f108c0c79bea8101 · report fixture: signature d_model:int/0/int, norm_type:str/0/str · returned RMSNorm · level recorded 2026-09-22
get_activation_function in 17 papers dual_attention_transformer.py ran · our draft was wrong MIT (permissive) · 074ed3e1882dc275 · report fixture: signature name:str/0/str · returned ReLU · level recorded 2026-09-22
DualAttention dual_attention_transformer.py unverified MIT (permissive) · 3cabca1c79c7e81e · report
DualAttnEncoderBlock dual_attention_transformer.py unverified MIT (permissive) · 85bb295631648984 · report
DualAttnTransformerLM dual_attention_transformer.py unverified MIT (permissive) · 81995fb31f7a5e45 · report

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it. Siblings are assemblies lifted from this paper in the same batch for the same method: alternative readings kept side by side; none replaces or supersedes another.

2 assemblies on this paper: 2 harvested, 0 generated, listed in 1 groups (siblings lifted in one batch form one group).

2 sibling assemblies for Dual Attention Transformer (DAT) (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for Dual Attention Transformer (DAT) (graph method: Dual Attention Transformer (DAT)). 18 bricks; individually on a fixture: 15 ran (of which 8 constructed an object rather than computing a result; 10 with no instrument failure: 1 honoured, 1 violated, 8 with no contract checked; 5 where Syntology's instrument failed) · 3 unverified. The assembly has not been run end-to-end.

one of 2 siblings

  1. DualAttnTransformerLM entry · level 0 · unverified · 81995fb31f7a5e45
  2. DualAttnEncoderBlock model · level 0 · unverified · 85bb295631648984
  3. PositionRelativeSymbolRetriever helper · level 1 · ran · 27aa5362b0778651
  4. PositionalSymbolRetriever helper · level 1 · ran · 07d2253837bc4d3f
  5. RelationalAttention model · level 1 · ran · f75b1ab91cff5f54
  6. SymbolicAttention model · level 1 · ran · 32699e73b61b32e9
  7. create_norm helper · level 1 · ran · our draft was wrong · f108c0c79bea8101
  8. precompute_freqs_cis helper · level 1 · ran · honoured contract · 5f447bdd807ed3c0
  9. DualAttention model · level 0 · unverified · 3cabca1c79c7e81e
  10. FeedForwardBlock model · level 1 · ran · a001e1fb0893402a
  11. RelativePositionalEncoding helper · level 1 · ran · 41fd4f0515e6826a
  12. apply_rotary_emb helper · level 1 · ran; the contract check could not run (RAISES) · 6dc1f541b73c3081
  13. compute_causal_mask helper · level 1 · ran · violated contract · a73cb1ba115d3af9
  14. get_activation_function helper · level 1 · ran · our draft was wrong · 074ed3e1882dc275
  15. repeat_kv helper · level 2 · ran; the contract check could not run · 6d2a08dcf3466514
  16. RMSNorm model · level 1 · ran · 01777e072296c0bb
  17. Attention model · level 1 · ran · 437880805ab90b88
  18. reshape_for_broadcast helper · level 1 · ran; the contract check could not run · e04ad9d4b02f50b6

harvested Assembly for Dual Attention Transformer (DAT) (graph method: Dual Attention Transformer (DAT)). 10 bricks; individually on a fixture: 6 ran (of which 1 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 1 violated, 1 with no contract checked; 4 where Syntology's instrument failed) · 4 unverified. The assembly has not been run end-to-end.

one of 2 siblings

  1. DualAttention entry · level 0 · unverified · 906b56a4d10f246c
  2. Attention helper · level 0 · unverified · ed6c6aa45ef23326
  3. DisentangledRelationalCrossAttention helper · level 0 · unverified · 65050ac2515e807a
  4. RelationalAttention helper · level 1 · ran · e96ef6f7fa7fb78b
  5. RelationalCrossAttention helper · level 0 · unverified · 8f6de9b1d93efd34
  6. apply_rotary_emb helper · level 1 · ran · our draft was wrong · dd15a61ea3d0e604
  7. compute_causal_mask helper · level 1 · ran · violated contract · a73cb1ba115d3af9
  8. repeat_kv helper · level 2 · ran; the contract check could not run · 6d2a08dcf3466514
  9. get_activation_function helper · level 1 · ran · our draft was wrong · 074ed3e1882dc275
  10. reshape_for_broadcast helper · level 1 · ran; the contract check could not run · e04ad9d4b02f50b6

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2405.16727") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Information RetrievalLanguage ModelingLanguage ModellingRelational ReasoningRetrieval

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Absolute Position EncodingsAdamAttentionBPEDense ConnectionsDropoutLabel SmoothingLayer NormalizationLinear LayerMulti-Head AttentionPosition-Wise Feed-Forward LayerResidual ConnectionSETSoftmaxTransformer

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