Papers › Breaking the Softmax Bottleneck: A High-Rank RNN Language Model
Breaking the Softmax Bottleneck: A High-Rank RNN Language Model
Zhilin Yang, Zihang Dai, Ruslan Salakhutdinov, William W. Cohen
We formulate language modeling as a matrix factorization problem, and show that the expressiveness of Softmax-based models (including the majority of neural language models) is limited by a Softmax bottleneck. Given that natural language is highly context-dependent, this further implies that in practice Softmax with distributed word embeddings does not have enough capacity to model natural language. We propose a simple and effective method to address this issue, and improve the state-of-the-art perplexities on Penn Treebank and WikiText-2 to 47.69 and 40.68 respectively. The proposed method also excels on the large-scale 1B Word dataset, outperforming the baseline by over 5.6 points in perplexity.
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):
get_harvested_code_for_paper(arxiv_id="1711.03953")get_code(code_sha256="…")with a sample'scode_sha256from the list belowget_citation_path(paper_1="1711.03953", paper_2="…")with another paper's arXiv id or titleget_concepts_for_paper(arxiv_id="1711.03953")
Code
Syntology 23 code samples harvested from 5 repositories linked to this paper: 12 ran (of which 0 constructed an object rather than computing a result; 11 with no instrument failure: 0 honoured, 0 violated, 11 with no contract checked; 1 where Syntology's instrument failed) · 11 unverified. Where Syntology's instrument failed: 1 where our contract draft was wrong; failures of Syntology's instrument, not of the code.
By repository: official repository: 3 samples from 1 repository, 2 ran (2 with no instrument failure; 0 where Syntology's instrument failed); community (archive-listed): 20 samples from 4 repositories, 10 ran (9 with no instrument failure; 1 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.
No sample is linked to this paper from 1 of its repositories that Syntology harvested for other papers:
- zhangyaoyuan/GAN-Simplification, community (archive-listed): harvested for another paper, listed under the repository
Levels written between 2026-09-22 and 2026-09-26 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.
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.
A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.
Code Syntology ran Syntology
23 samples harvested: 12 ran (of which 0 constructed an object rather than computing a result; 11 with no instrument failure: 0 honoured, 0 violated, 11 with no contract checked; 1 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 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.
Licence: 1 of the 23 samples is pointer only, meaning Syntology does not serve that copy's text. For that sample, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. This page shows no code text for any sample; each one links to its file in the repository.
Harvested from 5 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”. “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.
8585e034505891c4 · report
fixture: signature source:Tensor/1/float, i:int/0/int, args:Args/0/?, seq_len:NoneType/0/? · returned tuple · level recorded 2026-09-25 census: 5 implementations from 7 papers, 1 distinct output on one shared input dynamiceval.py ran MIT (permissive) ·
8bd3cf7e6979e441 · report
fixture: signature h:tuple/1/Tensor · returned tuple · level recorded 2026-09-25 612788fb8fc53b62 · report
6a3d8df0b754bb01 · report
census: 18 implementations from 25 papers, 10 distinct outputs across 6 buckets, one shared input per bucket awd/pointer.py unverified BSD-3-Clause (permissive) ·
8cfc87c86a68ccf9 · report
62876a934ad2732b · report
fixture: signature filename:str/0/str · returned dict · level recorded 2026-09-26 a2f146bb53bb255e · report
fixture: signature ref_path:str/0/str, hyp_path:str/0/str · returned tuple · level recorded 2026-09-26 73781c047e5bde5a · report
fixture: signature data_array:ndarray/1/float, _batch_size:int/0/int, _seq_len:int/0/int · returned tuple · level recorded 2026-09-26 1d654edced8659d4 · report
fixture: signature data_array:ndarray/1/float, batch_size:int/0/int, seq_len:int/0/int, shift:int/0/int · returned tuple · level recorded 2026-09-26 d78e2f03397f3d7c · report
fixture: signature data_array:ndarray/1/float, _batch_size:int/0/int, _seq_len:int/0/int, _shift:int/0/int, train:bool/0/bool · returned tuple · level recorded 2026-09-26 cbfd3e2359406dd6 · report
fixture: signature filename:str/0/str, word_to_id:dict/0/? · returned list · level recorded 2026-09-26 c98fe0c21eb27db8 · report
fixture: signature path:str/0/str · returned list · level recorded 2026-09-26 f5c5c321e38f81cd · report
fixture: signature filename:str/0/str · returned list · level recorded 2026-09-26 c75648e7493bfa87 · report
b47a3b26a9d39f5f · report
dd918220a90f8779 · report
7294b626ee6ccf20 · report
1720dba9c0088a7c · report
6006a3916540a413 · report
census: 6 implementations from 6 papers, 3 distinct outputs across 3 buckets, one shared input per bucket this sample: bucket 1 of 3, arg 1: rank 2, kind float, dtype float32 utils.py ran · our draft was wrong MIT (permissive) here; the stored copy is pointer only ·
2943dadcb85e6e3c · report
fixture: signature data:Tensor/2/float, bsz:int/0/int, args:Args/0/? · returned Tensor · level recorded 2026-09-22 591c3d72452ed508 · report
fixture: signature source:Tensor/2/float, i:int/0/int, args:Args/0/?, seq_len:NoneType/0/?, evaluation:bool/0/bool · returned tuple · level recorded 2026-09-26 61247fd8db2741cf · report
Tasks
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Language Modelling | Penn Treebank (Word Level) | AWD-LSTM-MoS + dynamic eval | Params | 22M | #11 of 43 | Archive leaderboard | report |
| Language Modelling | Penn Treebank (Word Level) | AWD-LSTM-MoS + dynamic eval | Test perplexity | 47.69 | #11 of 43 | Archive leaderboard | report |
| Language Modelling | Penn Treebank (Word Level) | AWD-LSTM-MoS + dynamic eval | Validation perplexity | 48.33 | #11 of 43 | Archive leaderboard | report |
| Language Modelling | Penn Treebank (Word Level) | AWD-LSTM-MoS | Params | 22M | #20 of 43 | Archive leaderboard | report |
| Language Modelling | Penn Treebank (Word Level) | AWD-LSTM-MoS | Test perplexity | 54.44 | #20 of 43 | Archive leaderboard | report |
| Language Modelling | Penn Treebank (Word Level) | AWD-LSTM-MoS | Validation perplexity | 56.54 | #20 of 43 | Archive leaderboard | report |
| Language Modelling | WikiText-2 | AWD-LSTM-MoS + dynamic eval | Number of params | 35M | #16 of 38 | Archive leaderboard | report |
| Language Modelling | WikiText-2 | AWD-LSTM-MoS + dynamic eval | Test perplexity | 40.68 | #16 of 38 | Archive leaderboard | report |
| Language Modelling | WikiText-2 | AWD-LSTM-MoS + dynamic eval | Validation perplexity | 42.41 | #16 of 38 | Archive leaderboard | report |
| Language Modelling | WikiText-2 | AWD-LSTM-MoS | Number of params | 35M | #26 of 38 | Archive leaderboard | report |
| Language Modelling | WikiText-2 | AWD-LSTM-MoS | Test perplexity | 61.45 | #26 of 38 | Archive leaderboard | report |
| Language Modelling | WikiText-2 | AWD-LSTM-MoS | Validation perplexity | 63.88 | #26 of 38 | Archive leaderboard | report |
Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.
Methods
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