Papers › Pre-Trained Language Models for Interactive Decision-Making

Pre-Trained Language Models for Interactive Decision-Making

3 Feb 2022arXiv:2202.01771archive 2025-07-28

Shuang Li, Xavier Puig, Chris Paxton, Yilun Du, Clinton Wang, Linxi Fan, Tao Chen, De-An Huang, Ekin Akyürek, Anima Anandkumar, Jacob Andreas, Igor Mordatch, Antonio Torralba, Yuke Zhu

Language model (LM) pre-training is useful in many language processing tasks. But can pre-trained LMs be further leveraged for more general machine learning problems? We propose an approach for using LMs to scaffold learning and generalization in general sequential decision-making problems. In this approach, goals and observations are represented as a sequence of embeddings, and a policy network initialized with a pre-trained LM predicts the next action. We demonstrate that this framework enables effective combinatorial generalization across different environments and supervisory modalities. We begin by assuming access to a set of expert demonstrations, and show that initializing policies with LMs and fine-tuning them via behavior cloning improves task completion rates by 43.6% in the VirtualHome environment. Next, we integrate an active data gathering procedure in which agents iteratively interact with the environment, relabel past "failed" experiences with new goals, and update their policies in a self-supervised loop. Active data gathering further improves combinatorial generalization, outperforming the best baseline by 25.1%. Finally, we explain these results by investigating three possible factors underlying the effectiveness of the LM-based policy. We find that sequential input representations (vs. fixed-dimensional feature vectors) and LM-based weight initialization are both important for generalization. Surprisingly, however, the format of the policy inputs encoding (e.g. as a natural language string vs. an arbitrary sequential encoding) has little influence. Together, these results suggest that language modeling induces representations that are useful for modeling not just language, but also goals and plans; these representations can aid learning and generalization even outside of language processing.

PaperPDFCodeCode Syntology ran

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: 13 ran (of which 0 constructed an object rather than computing a result; 13 with no instrument failure: 0 honoured, 0 violated, 13 with no contract checked; 0 where Syntology's instrument failed) · 5 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code; here there are none.

By repository: community (archive-listed): 18 samples from 1 repository, 13 ran (13 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 2026-09-24 and 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

xavierpuigf/virtualhome mentioned on GitHubMITSyntology: 18 samples, 13 ran (0 instrument-failed) · 5 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

18 samples harvested: 13 ran (of which 0 constructed an object rather than computing a result; 13 with no instrument failure: 0 honoured, 0 violated, 13 with no contract checked; 0 where Syntology's instrument failed) · 5 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.

13ran
5unverified

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 xavierpuigf/virtualhome. “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.

xavierpuigf/virtualhome community (archive-listed) 18 samples, 13 ran (0 instrument-failed) · 5 unverified
build_grid_images virtualhome/demo/generate_snapshot.py ran MIT (permissive) · adbcc83fbce194cc · report fixture: signature images:list/1/list · returned ndarray · level recorded 2026-09-24
clean_graph virtualhome/demo/utils_demo.py ran MIT (permissive) · 83295b4614f48186 · report fixture: signature graph:dict/0/? · returned dict · level recorded 2026-09-24
cuboid virtualhome/demo/display_graph.py ran MIT (permissive) · b3ca8a1b900d75f3 · report fixture: signature ax:Axes3D/0/?, coords:list/1/float, sides:list/1/float · returned Poly3DCollection · level recorded 2026-09-25
find_edges_from virtualhome/demo/utils_demo.py ran MIT (permissive) · 44dc4121fc6621b0 · report fixture: signature graph:dict/0/?, id:int/0/int · returned list · level recorded 2026-09-24
from_hash virtualhome/dataset_utils/augment_dataset_exceptions.py ran MIT (permissive) · 058686fb222c2d31 · report fixture: signature precond_tuple:tuple/1/tuple · returned list · level recorded 2026-09-24
getidperobject virtualhome/dataset_utils/exception_handler.py ran MIT (permissive) · 05083186903d3268 · report fixture: signature object_name:str/0/str, id_env:int/0/int, id_mapping:dict/0/? · returned int · level recorded 2026-09-24
insertInstructions virtualhome/dataset_utils/augmentation_utils.py ran MIT (permissive) · f44fbe71aa27eb3d · report fixture: signature insert_in:list/1/tuple, contentold:list/1/str · returned list · level recorded 2026-09-24
obtain_objects_from_message virtualhome/dataset_utils/execute_script_utils.py ran MIT (permissive) · b0f5c4fe4cd3952f · report fixture: signature message:dict/0/? · returned list · level recorded 2026-09-24
obtain_scene_id_from_path virtualhome/dataset_utils/execute_script_utils.py ran MIT (permissive) · eb99b8eb5c8721af · report fixture: signature path:str/0/str · returned int · level recorded 2026-09-24
obtain_script_grounded_in_graph virtualhome/dataset_utils/augment_dataset_exceptions.py ran MIT (permissive) · 287b61eceef14fa4 · report fixture: signature lines_program:list/1/str, id_mapping:dict/0/?, modified_script:list/1/ScriptLine · returned list · level recorded 2026-09-24
parseStrBlock virtualhome/dataset_utils/augmentation_utils.py ran fingerprinted MIT (permissive) · 6b36626f01871673 · report fixture: signature block_str:str/0/str · returned tuple · level recorded 2026-09-24
parse_exec_script_file virtualhome/dataset_utils/execute_script_utils.py ran MIT (permissive) · f5247aeac1223674 · report fixture: signature file_name:str/0/str · returned tuple · level recorded 2026-09-24
to_hash virtualhome/dataset_utils/augment_dataset_exceptions.py ran MIT (permissive) · 998d6b9491c17a70 · report fixture: signature precond_list:list/1/dict · returned tuple · level recorded 2026-09-24
find_nodes virtualhome/demo/utils_demo.py unverified MIT (permissive) · 8e3413a872edc331 · report
get_preconds_script virtualhome/dataset_utils/add_preconds.py unverified MIT (permissive) · 1aef779a0ec40326 · report
get_xyz_mouse_click virtualhome/demo/display_graph.py unverified MIT (permissive) · fc264d88eed3353e · report
hasProperty virtualhome/dataset_utils/augmentation_utils.py unverified MIT (permissive) · ff70381f7e6ddae8 · report
parseException virtualhome/dataset_utils/exception_handler.py unverified MIT (permissive) · b58216b38e8b12f8 · report

Tasks

Decision MakingImitation LearningLanguage ModelingLanguage ModellingSequential Decision Making

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

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