@inproceedings{aroca-ouellette-etal-2025-reseeding,
title = "{R}e{S}eeding Latent States for Sequential Language Understanding",
author = "Aroca-Ouellette, St{\'e}phane and
von der Wense, Katharina and
Roncone, Alessandro",
editor = "Christodoulopoulos, Christos and
Chakraborty, Tanmoy and
Rose, Carolyn and
Peng, Violet",
booktitle = "Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2025",
address = "Suzhou, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.emnlp-main.1281/",
pages = "25233--25247",
ISBN = "979-8-89176-332-6",
abstract = "We introduce Refeeding State Embeddings aligned using Environmental Data (ReSEED), a novel method for grounding language in environmental data. While large language models (LLMs) excel at many tasks, they continue to struggle with multi-step sequential reasoning. ReSEED addresses this by producing latent embeddings aligned with the true state of the environment and refeeding these embeddings into the model before generating its output. To evaluate its effectiveness, we develop three new sequential reasoning benchmarks, each with a training set of paired state-text trajectories and several text-only evaluation sets that test generalization to longer, unseen trajectories. Across all benchmarks, ReSEED significantly improves generalization and scalability over a text-only baseline. We further show that ReSEED outperforms commercial LLMs on our benchmarks, highlighting the value of grounding language in the environment."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="aroca-ouellette-etal-2025-reseeding">
<titleInfo>
<title>ReSeeding Latent States for Sequential Language Understanding</title>
</titleInfo>
<name type="personal">
<namePart type="given">Stéphane</namePart>
<namePart type="family">Aroca-Ouellette</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Katharina</namePart>
<namePart type="family">von der Wense</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alessandro</namePart>
<namePart type="family">Roncone</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Christos</namePart>
<namePart type="family">Christodoulopoulos</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tanmoy</namePart>
<namePart type="family">Chakraborty</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Carolyn</namePart>
<namePart type="family">Rose</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Violet</namePart>
<namePart type="family">Peng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Suzhou, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-332-6</identifier>
</relatedItem>
<abstract>We introduce Refeeding State Embeddings aligned using Environmental Data (ReSEED), a novel method for grounding language in environmental data. While large language models (LLMs) excel at many tasks, they continue to struggle with multi-step sequential reasoning. ReSEED addresses this by producing latent embeddings aligned with the true state of the environment and refeeding these embeddings into the model before generating its output. To evaluate its effectiveness, we develop three new sequential reasoning benchmarks, each with a training set of paired state-text trajectories and several text-only evaluation sets that test generalization to longer, unseen trajectories. Across all benchmarks, ReSEED significantly improves generalization and scalability over a text-only baseline. We further show that ReSEED outperforms commercial LLMs on our benchmarks, highlighting the value of grounding language in the environment.</abstract>
<identifier type="citekey">aroca-ouellette-etal-2025-reseeding</identifier>
<location>
<url>https://aclanthology.org/2025.emnlp-main.1281/</url>
</location>
<part>
<date>2025-11</date>
<extent unit="page">
<start>25233</start>
<end>25247</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T ReSeeding Latent States for Sequential Language Understanding
%A Aroca-Ouellette, Stéphane
%A von der Wense, Katharina
%A Roncone, Alessandro
%Y Christodoulopoulos, Christos
%Y Chakraborty, Tanmoy
%Y Rose, Carolyn
%Y Peng, Violet
%S Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing
%D 2025
%8 November
%I Association for Computational Linguistics
%C Suzhou, China
%@ 979-8-89176-332-6
%F aroca-ouellette-etal-2025-reseeding
%X We introduce Refeeding State Embeddings aligned using Environmental Data (ReSEED), a novel method for grounding language in environmental data. While large language models (LLMs) excel at many tasks, they continue to struggle with multi-step sequential reasoning. ReSEED addresses this by producing latent embeddings aligned with the true state of the environment and refeeding these embeddings into the model before generating its output. To evaluate its effectiveness, we develop three new sequential reasoning benchmarks, each with a training set of paired state-text trajectories and several text-only evaluation sets that test generalization to longer, unseen trajectories. Across all benchmarks, ReSEED significantly improves generalization and scalability over a text-only baseline. We further show that ReSEED outperforms commercial LLMs on our benchmarks, highlighting the value of grounding language in the environment.
%U https://aclanthology.org/2025.emnlp-main.1281/
%P 25233-25247
Markdown (Informal)
[ReSeeding Latent States for Sequential Language Understanding](https://aclanthology.org/2025.emnlp-main.1281/) (Aroca-Ouellette et al., EMNLP 2025)
ACL
- Stéphane Aroca-Ouellette, Katharina von der Wense, and Alessandro Roncone. 2025. ReSeeding Latent States for Sequential Language Understanding. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 25233–25247, Suzhou, China. Association for Computational Linguistics.