Papers › One Model, Two Roles: Emergent Specialization in a Shared Recurrent Transformer

One Model, Two Roles: Emergent Specialization in a Shared Recurrent Transformer

18 May 2026arXiv:2605.17811added by Syntology

Jucheng Shen, Barbara Su, Anastasios Kyrillidis

Title, abstract, authors and date from arXiv's metadata (CC0); this paper is not in the Papers with Code archive (frozen 2025-07-28).

Can a shared-weight recurrent Transformer develop distinct internal roles without being partitioned into separate modules? We study this in Asymmetric Input Recurrence (AIR), a minimal two-state reasoning architecture in which the same Transformer model is reused for both updates (per literature, L and H) and the only built-in difference in the update rule is that the encoded input is injected during L-updates but not H-updates. Across Sudoku-Extreme and Maze, decoded rollouts reveal a consistent split: zH behaves like a fully committed proposal state, whereas zL retains local uncertainty and shifting intermediate structure. Freeze experiments show that this split is, in practice, related to the model's state dynamics: in Sudoku, freezing zH reduces zL's content changes whereas freezing zL increases zH's, while in Maze, freezing either state increases content changes in the other state. Ablations show that to induce specialization, the shared model needs to be able to tell the two update types apart, either from input injection asymmetry or from a separate level token. Mechanistically, attention analysis shows that L-updates are consistently more local than H-updates in both Sudoku and Maze. Together, these results show that, in a two-state recurrent setting, a clear state-identity signal can induce stable, related functional roles inside a shared-parameter recurrent Transformer. Code is available at https://github.com/juchengshen/air.

PaperPDFCode 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 18 code samples harvested from 1 repository linked to this paper: 7 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 0 violated, 3 with no contract checked; 3 where Syntology's instrument failed) · 11 unverified. Where Syntology's instrument failed: 2 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: found in paper text by Syntology: 18 samples from 1 repository, 7 ran (4 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.

juchengshen/air found in paper text by SyntologySyntology: 18 samples, 7 ran (3 instrument-failed) · 11 unverified report

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

18 samples harvested: 7 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 0 violated, 3 with no contract checked; 3 where Syntology's instrument failed) · 11 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · honoured contract
2ran · our draft was wrong
1ran; the contract check could not run
3ran
11unverified

Licence: 0 of the 18 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 juchengshen/air. “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”. “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.

juchengshen/air found in paper text by Syntology 18 samples, 7 ran (3 instrument-failed) · 11 unverified
CastedEmbedding models/air/air_1net_L2x_H2x_input_token_prepend.py ran · metamorphic tier: invariant Apache-2.0 (permissive) · c21ab832360c7a4e · report fixture: signature input:Tensor/1/int · returned Tensor · level recorded 2026-09-22
CastedLinear models/air/air_1net_L2x_H2x_input_token_prepend.py ran · metamorphic tier: invariant Apache-2.0 (permissive) · 5e1bae41e0f0de3b · report fixture: signature input:Tensor/2/float · returned Tensor · level recorded 2026-09-22
SwiGLU in 9 papers models/air/air_1net_L2x_H2x_input_token_prepend.py ran · metamorphic tier: invariant Apache-2.0 (permissive) · 0160973ed66c95fd · report fixture: signature x:Tensor/3/float · returned Tensor · level recorded 2026-09-22
_find_multiple models/air/air_1net_L2x_H2x_input_token_prepend.py ran · honoured contract fingerprinted Apache-2.0 (permissive) · ee6b213d67acc21f · report fixture: signature a:int/0/int, b:int/0/int · returned int · level recorded 2026-09-22
rms_norm in 10 papers models/air/air_1net_L2x_H2x_input_token_prepend.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 277cf09c927bf2c0 · report fixture: signature hidden_states:Tensor/3/float, variance_epsilon:float/0/float · returned Tensor · level recorded 2026-09-22
rotate_half in 424 papers
census: 24 implementations from 96 papers, 4 distinct outputs across 2 buckets, one shared input per bucket this sample: bucket 1 of 2, arg 1: rank 2, kind float, dtype float32
models/air/air_1net_L2x_H2x_input_token_prepend.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 5d0ae16200be190c · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-22
trunc_normal_init_ models/air/air_1net_L2x_H2x_input_token_prepend.py ran; the contract check could not run (RAISES) fingerprinted Apache-2.0 (permissive) · e9e39505360d7d07 · report fixture: signature tensor:Tensor/2/float, std:float/0/float, lower:float/0/float, upper:float/0/float · returned Tensor · level recorded 2026-09-22
AIR models/air/air_1net_L2x_H2x_input_token_prepend.py unverified Apache-2.0 (permissive) · 5576aa66b9ec2ea4 · report
AIRBlock models/air/air_1net_L2x_H2x_input_token_prepend.py unverified Apache-2.0 (permissive) · 1df5376995a79795 · report
AIRCarry models/air/air_1net_L2x_H2x_input_token_prepend.py unverified Apache-2.0 (permissive) · 6bb1e0108ac99fa1 · report
AIRConfig models/air/air_1net_L2x_H2x_input_token_prepend.py unverified Apache-2.0 (permissive) · a012233c3527ade7 · report
AIRInnerCarry models/air/air_1net_L2x_H2x_input_token_prepend.py unverified Apache-2.0 (permissive) · 55ce9b34d823aa4d · report
AIRReasoningModule models/air/air_1net_L2x_H2x_input_token_prepend.py unverified Apache-2.0 (permissive) · e2fa7ed580a7f6f9 · report
AIR_Inner models/air/air_1net_L2x_H2x_input_token_prepend.py unverified Apache-2.0 (permissive) · 335ba801be5558c2 · report
CastedSparseEmbedding models/air/air_1net_L2x_H2x_input_token_prepend.py unverified Apache-2.0 (permissive) · 9d3cb2d57bdf4030 · report
RotaryEmbedding in 27 papers models/air/air_1net_L2x_H2x_input_token_prepend.py unverified Apache-2.0 (permissive) · ecf65cb1949cf62e · 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.

1 assembly on this paper: 1 harvested, 0 generated.

harvested Assembly for Asymmetric Input Recurrence (AIR) (graph method: Asymmetric Input Recurrence (AIR)). 18 bricks; individually on a fixture: 7 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 0 violated, 3 with no contract checked; 3 where Syntology's instrument failed) · 11 unverified. The assembly has not been run end-to-end.

  1. AIR entry · level 0 · unverified · 5576aa66b9ec2ea4
  2. AIRCarry model · level 0 · unverified · 6bb1e0108ac99fa1
  3. AIRConfig helper · level 0 · unverified · a012233c3527ade7
  4. AIR_Inner model · level 0 · unverified · 335ba801be5558c2
  5. AIRInnerCarry model · level 0 · unverified · 55ce9b34d823aa4d
  6. AIRBlock model · level 0 · unverified · 1df5376995a79795
  7. AIRReasoningModule model · level 0 · unverified · e2fa7ed580a7f6f9
  8. CastedEmbedding helper · level 1 · ran · c21ab832360c7a4e
  9. CastedLinear helper · level 1 · ran · 5e1bae41e0f0de3b
  10. CastedSparseEmbedding helper · level 0 · unverified · 9d3cb2d57bdf4030
  11. RotaryEmbedding helper · level 0 · unverified · ecf65cb1949cf62e
  12. trunc_normal_init_ helper · level 1 · ran; the contract check could not run (RAISES) · e9e39505360d7d07
  13. Attention helper · level 0 · unverified · e1fe279de6f03a37
  14. SwiGLU helper · level 1 · ran · 0160973ed66c95fd
  15. rms_norm helper · level 1 · ran · our draft was wrong · 277cf09c927bf2c0
  16. apply_rotary_pos_emb helper · level 0 · unverified · 99138bc8ae291d5f
  17. _find_multiple helper · level 1 · ran · honoured contract · ee6b213d67acc21f
  18. rotate_half helper · level 1 · ran · our draft was wrong · 5d0ae16200be190c

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="2605.17811") returns these assemblies with each brick's role and wiring (how to connect).

Results from the paper

The Papers with Code archive ends with its 2025-07-28 snapshot. This paper's arXiv identifier, 2605.17811, was issued in May 2026, after that date, so the archive has no leaderboard rows for it.

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