Papers › Contrastive Reinforcement Learning of Symbolic Reasoning Domains

Contrastive Reinforcement Learning of Symbolic Reasoning Domains

16 Jun 2021NeurIPS 2021 12arXiv:2106.09146archive 2025-07-28

Gabriel Poesia, WenXin Dong, Noah Goodman

Abstract symbolic reasoning, as required in domains such as mathematics and logic, is a key component of human intelligence. Solvers for these domains have important applications, especially to computer-assisted education. But learning to solve symbolic problems is challenging for machine learning algorithms. Existing models either learn from human solutions or use hand-engineered features, making them expensive to apply in new domains. In this paper, we instead consider symbolic domains as simple environments where states and actions are given as unstructured text, and binary rewards indicate whether a problem is solved. This flexible setup makes it easy to specify new domains, but search and planning become challenging. We introduce four environments inspired by the Mathematics Common Core Curriculum, and observe that existing Reinforcement Learning baselines perform poorly. We then present a novel learning algorithm, Contrastive Policy Learning (ConPoLe) that explicitly optimizes the InfoNCE loss, which lower bounds the mutual information between the current state and next states that continue on a path to the solution. ConPoLe successfully solves all four domains. Moreover, problem representations learned by ConPoLe enable accurate prediction of the categories of problems in a real mathematics curriculum. Our results suggest new directions for reinforcement learning in symbolic domains, as well as applications to mathematics education.

PaperPDFConference PDFCodeCode 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.

Syntology holds a reference implementation of Contrastive Policy Learning (ConPoLe), generated from the paper text (not the authors' code), at level 3 on Syntology's ladder: V2 “two implementations GENERATED from the paper's signature and spec agree numerically (1e-6) on fixture inputs” “neither of the two ever read the authors' own code, so V2 measures how unambiguously the paper specifies the method, not fidelity to the authors' implementation (measured 2026-09-11, commit 75eaa77: cross_checked_by is a second LLM on 2,628 of 2,831 served samples)”; V3 “paper-derived property tests pass on both of those same two generations.” (quoted from Syntology's server card). No rung compares this code against the authors' implementation. For agents: get_reference_implementation(name='Contrastive Policy Learning (ConPoLe)') (how to connect); the free check have('2106.09146') needs no token.

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

Code

Syntology 15 code samples harvested from 2 repositories linked to this paper: 4 ran (of which 1 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 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: 14 samples from 1 repository, 3 ran (2 with no instrument failure; 1 where Syntology's instrument failed); community (archive-listed): 1 sample from 1 repository, 1 ran (1 with no instrument failure; 0 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 between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

gpoesia/socratic-tutor officialmentioned in papermentioned on GitHubpytorchSyntology: 14 samples, 3 ran (1 instrument-failed) · 11 unverified report
sagnikanupam/mathdsl mentioned on GitHubpytorchSyntology: 1 sample, 1 ran (0 instrument-failed) · 0 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

15 samples harvested: 4 ran (of which 1 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 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.

1ran · our draft was wrong
3ran
11unverified

Licence: 15 of the 15 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”. “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.

gpoesia/socratic-tutor official repository 14 samples, 3 ran (1 instrument-failed) · 11 unverified
RacketEnvironment agent.py ran no licence file found · pointer only · 3d82da901d88aa3c · report fixture: signature states:list/1/State, domain:str/0/str · returned list · level recorded 2026-09-24
random_initial_seed agent.py ran no licence file found · pointer only · 8e74322415ba2a24 · report fixture: returned int · level recorded 2026-09-24
register in 66 papers agent.py ran · our draft was wrong no licence file found · pointer only · 6a1ad0909c55aa20 · report fixture: signature superclass:callable/0/? · returned function · level recorded 2026-09-22
Action in 10 papers agent.py unverified no licence file found · pointer only · 82b383962990e9ba · report
ContrastiveExample agent.py unverified no licence file found · pointer only · e38eefb547b2a7fd · report
InverseLength agent.py unverified no licence file found · pointer only · bbc1dc4497b5ec6b · report
LearningAgent agent.py unverified no licence file found · pointer only · 6740a0d260d2a467 · report
MultiTaskEnvironment agent.py unverified no licence file found · pointer only · cd47224c2bdbef51 · report
NCE agent.py unverified no licence file found · pointer only · 519728cf68d959ec · report
QFunction agent.py unverified no licence file found · pointer only · 05426ee7aeefc5ac · report
RandomQFunction agent.py unverified no licence file found · pointer only · 4566b327bd686f0c · report
RubiksGreedyHeuristic agent.py unverified no licence file found · pointer only · 2db358be2ba65b5e · report
RustEnvironment agent.py unverified no licence file found · pointer only · 6f6ab6bad7d94b64 · report
State in 13 papers agent.py unverified no licence file found · pointer only · a263044fe206c128 · report
sagnikanupam/mathdsl community (archive-listed) 1 sample, 1 ran (0 instrument-failed) · 0 unverified
Tree in 10 papers dreamcoder/domains/conpole/conpoleMathPrimitives.py ran no licence file found · pointer only · 43eefe3507985e75 · report fixture: returned Tree · level recorded 2026-09-22

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. For a generated assembly, an independent peer is a second implementation generated separately from the same paper text; neither implementation read the authors' code. At level 2 the two agreed numerically on the fixture's inputs; level 3 is recorded when property tests derived from the paper also passed on both, and each level 3 card prints how many its report lists. 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. A generated assembly's label was set when its code was generated for the method, and no model judged it; its card prints it as "by construction" or "not judged by a model". 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. Where an assembly's phrase has no constructed figure, a brick that ran is not among the harvested samples this build reads, and whether a brick's run only constructed an object is read from that brick's harvested sample, so the figure was not counted; it does not say there were none.

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

generated Assembly for Contrastive Policy Learning (ConPoLe) (graph method: Contrastive Policy Learning (ConPoLe)), generated from the paper text, not the repository's code. 1 brick; individually on a fixture: 1 ran (1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified. Ran end-to-end against an independent peer (level 3; 5 property tests passed on both).

label matches by construction: the code was generated for this method (not judged by a model)

  1. info_nce_loss entry · level 3 · ran · 14c87df619a1cb9b236315acf61b497b84f7a0c97c0551c7354ab7d36d2c8049

2 sibling assemblies for Contrastive Policy Learning (ConPoLe) (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for Contrastive Policy Learning (ConPoLe) (graph method: Contrastive Policy Learning (ConPoLe)). 1 brick; individually on a fixture: 1 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified; the one sample that ran constructed an object rather than computing a result. The assembly has not been run end-to-end.

one of 2 siblings

  1. Tree entry · level 1 · ran · 43eefe3507985e75

harvested Assembly for Contrastive Policy Learning (ConPoLe) (graph method: Contrastive Policy Learning (ConPoLe)). 14 bricks; individually on a fixture: 3 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 1 where Syntology's instrument failed) · 11 unverified. The assembly has not been run end-to-end.

one of 2 siblings

  1. NCE entry · level 0 · unverified · 519728cf68d959ec
  2. ContrastiveExample helper · level 0 · unverified · e38eefb547b2a7fd
  3. InverseLength helper · level 0 · unverified · bbc1dc4497b5ec6b
  4. LearningAgent helper · level 0 · unverified · 6740a0d260d2a467
  5. RandomQFunction helper · level 0 · unverified · 4566b327bd686f0c
  6. RubiksGreedyHeuristic helper · level 0 · unverified · 2db358be2ba65b5e
  7. register helper · level 1 · ran · our draft was wrong · 6a1ad0909c55aa20
  8. Action helper · level 0 · unverified · 82b383962990e9ba
  9. QFunction helper · level 0 · unverified · 05426ee7aeefc5ac
  10. State helper · level 0 · unverified · a263044fe206c128
  11. random_initial_seed helper · level 1 · ran · 8e74322415ba2a24
  12. MultiTaskEnvironment helper · level 0 · unverified · cd47224c2bdbef51
  13. RacketEnvironment helper · level 1 · ran · 3d82da901d88aa3c
  14. RustEnvironment helper · level 0 · unverified · 6f6ab6bad7d94b64

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="2106.09146") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Reinforcement LearningReinforcement Learning (RL)reinforcement-learning

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

InfoNCE

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