United States

Most existing work on mental health predic-
tion from language focuses on isolated posts,

overlooking temporal dynamics in longitudinal

timelines. We present McMaster NLP’s sys-
tem for the CLPsych 2026 Shared Task, which

centers on modeling mental health dynamics in

social media timelines using the MIND frame-
work (Atzil-Slonim, 2025). The task com-
prises: (1) identifying adaptive and maladap-
tive self-state components within posts, (2) de-
tecting moments of change in wellbeing, and

(3) generating structured summaries. For self-
state prediction, we leverage LLM-generated

archetypal representations of language use as

semantic anchors within a dual-encoder archi-
tecture, enabling interpretable prediction of

subelements and their intensities through align-
ment with prototypical expressions of psycho-
logical states. For temporal dynamics, we

use BiLSTM-based sequence models to de-
tect moments of change. For summarization,

we employ a prompt-based LLM to generate
grounded, structured summaries emphasizing
causal interactions and temporal progression of
self-states. Finally, we analyze model failure
modes with respect to human evaluation and
identify directions for reconciling the MIND
framework with how state-assessment models
encode meaning.

ACL 2026

McMasters of Change: Predicting Wellbeing States and Transitions from Longitudinal Language

bilstm-based change detection

abcd archetype similarity features

mentalroberta embeddings

### Welcome to 2025 Annual Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics

Welcome to the 2025 meeting of the Nations of the Americas Chapter of the Association for Computational Linguistics! I am proud to help organize the first NAACL conference to carry the new name of our organization, one that emphasizes inclusion for all of the Americas. I am also pleased to welcome you to Albuquerque, New Mexico, a state whose unique blend of cultural influences will make for an excellent backdrop for NAACL 2025, especially with this year’s special theme on NLP in a Multicultural World. 
**[Continue reading...](https://drive.google.com/file/d/1jX-qGhqVSZZCIrAnJaz798pflu5Irrdn/view?usp=sharing)**

*- Colin Cherry, Google, NAACL 2025 General Chair* 

[![](https://assets.underline.io/markdown_image/1/image/b087f8a4dc5816d6e1a6514e59c59ac3.png)](https://drive.google.com/file/d/1T96GzPqObXrMln2BMByCSXTSizjTg69P/view?usp=sharing)

You need to log in with the email address you registered with. Access credentials have been sent to your email. 

Please be sure to check your spam and other email folders if you do not see an email confirmation right away.

Please log in to explore this event.

To access NAACL 2025 event page you are required to register. Please follow [**this link**](https://2025.naacl.org/registration/registration/) to register. Access will depend on your registration type.

Please register!

NAACL 2025

Welcome to 2025 Annual Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics

workshop paper

It is my great pleasure and honor to welcome you to the 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026), held in San Diego, California, USA, July 2–7, 2026. ACL 2026 continues our fields tradition of excellence in scholarship, innovation, and inclusivity, and I am deeply grateful to the many volunteers who have worked tirelessly to bring this event to life.

It is exciting to see the many developments in the field reflected at our conference, such as the advancement of large language models (born out of our community) into many application directions. This is not without challenges, partly due to unsolved problems, and partly due to misuse of the technology. All this is reflected well at ACL 2026. The conference also reflects that the field is not only about advancing technology, but also the science of fundamentally understanding language and how it is use by many different speaker communities. 
[Full Message from General Chair Philipp Koehn](https://drive.google.com/file/d/1QiJFOmMjvDgsDo3i1XzlQnImxe3RU_Cy/view?usp=sharing) 
<html><button style="display: inline-flex; align-items: center; justify-content: center; white-space: nowrap; border-radius: 9999px; font-weight: bold; background: #7c3aed; color: white; font-family: 'Space Grotesk', sans-serif; height: 40px; font-size: 16px; padding: 0 20px; border: none; cursor: pointer" onclick="window.open('https://drive.google.com/file/d/1QiJFOmMjvDgsDo3i1XzlQnImxe3RU_Cy/view?usp=sharing')">Conference Handbook</button></html>

To access this event page, you need to log in with the **email address you registered with**. Access credentials will be sent to your email from Underline - subject line "Welcome to ACL 2026". Please be sure to check your spam email folder if you do not see an email confirmation right away.

To access this event page, you are required to register.
Please complete your registration to continue.

We recommend reading [**the registration information**](https://2026.aclweb.org/registration/) first.

Registration Required

The 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026) will take place in San Diego, California, from July 2 - 7, 2026.

This paper presents a system for the CLPsych
2026 Shared Task on longitudinal mental
health modeling from social media timelines,

grounded in the MIND framework (Atzil-
Slonim, 2025). MIND conceptualizes mental

health as evolving self-states defined by Affect,

Behavior, Cognition, and Desire (ABCD), pro-
viding a structured lens on mental health trajec-
tories. The system centers on a theory-explicit

prompting framework for structured sequence

summarization (Task 3.1) and recurrent dy-
namic signature extraction (Task 3.2), encoding

the full ABCD taxonomy directly into the LLM

prompt to ensure clinically grounded, inter-
pretable outputs. A three-stage pipeline infers

a direction-of-change label per sequence, pro-
duces structured ABCD summaries with few-
shot exemplar augmentation, and aggregates

these summaries to derive cross-individual re-
current patterns. The system ranks first on

deterioration-related recurrent signatures and

second overall, achieving the top Fit and Speci-
ficity scores in Task 3.2, demonstrating the ben-
efits of explicit clinical grounding for concep-
tual accuracy.

Theory-Explicit Prompting for MIND Self-States: Hierarchical LLMs and Dynamic Signature Extraction in Mental Health Timelines

This paper describes the StateOfMIND team
submission to Task 1 of the CLPsych 2026

shared task on identifying adaptive and mal-
adaptive psychological self-states in social me-
dia posts. For each post, the task requires

systems to predict which fine-grained subele-
ments of a self-state are expressed (Task 1.1)

and to rate how strongly adaptive and maladap-
tive states are present (Task 1.2). We address

both subtasks with a retrieval-augmented in-
context learning ensemble of two open-weight

LLMs, Qwen3.5-27B and Mistral-Small-3.2-

24B-Instruct. Our approach combines label-
aware retrieval of in-context examples, a three-
prompt decomposition targeting adaptive con-
tent and the Affect element, and subtask-
specific ensemble rules that favor precision for

subelement classification and recall for pres-
ence rating. Our best submissions rank 2nd

of 17 teams on Task 1.1 (Macro F1 = 0.441)
and 5th of 17 teams on Task 1.2 (RMSE =

0.994), with ablations showing retrieval and en-
sembling as the largest drivers of performance.

Self-State Identification with Retrieved In-Context Examples and Open-Weight LLMs

We describe the Meronym Labs system for
the CLPsych 2026 shared task. Our approach
uses retrieval-augmented in-context learning
with frozen LLMs throughout, requiring no
fine-tuning. Tasks 1 and 2 use a single
Qwen 3.5 27B call per post/timeline with static
and dynamically retrieved examples. Task 3.1
uses a five-agent agentic pipeline including

rule-based agents for change type and direc-
tion, LLM agents for dynamics extraction

and summary writing, and a validator all aug-
mented with NLI-based candidate re-ranking

and iterative contradiction reduction. The sys-
tem ranked 1st on Task 1.2 (RMSE 0.917)

and Task 3.1 (score rank average 4.00), 3rd
on Task 1.1 (F1 0.420), and 8th on Task 2
(F1 0.466).

Agentic Pipelines Meet Retrieval-Augmented ICL: A Zero-Training Approach to Mental Health Modeling

The critique of scalar benchmark rankings as proxies for model quality is now well-established (Raji et al., 2021; Wallach et al., 2025; Bean et al., 2025; Gehrmann et al., 2021). What the field still lacks is a shared structural vocabulary for comparing, combining, and contextualizing metric design choices. This paper provides that vocabulary: a four-primitive typology---representation ($\phi$), comparison ($D$), aggregation ($A$), and context ($C$)---under which existing metrics (BLEU, BERTScore, nDCG, LLM-as-judge, calibration scores, agentic outcome measures) are explicit parameterizations of a common form. This typology is paired with a measurement--decision split: metrics are noisy estimators of latent constructs, and model selection is context-dependent Pareto optimization over construct estimates, not over raw scores. The typology makes implicit metric assumptions comparable and debatable rather than hidden inside a single number.

Toward a Metric Typology for Language Model Evaluation

Recent advances in large language models (LLMs) have demonstrated impressive medical reasoning capabilities. However, current evaluation methods are mostly limited to static case vignettes and multiple-choice questions which fail to reflect the complexity, uncertainty, and iterative nature of real-world clinical decision-making. To bridge this gap, we propose **DiagBench**, a novel benchmark where models interact dynamically with a LLM based Patient Simulator, querying relevant clinical details to formulate accurate diagnoses. To complement this, we introduce **MedConvBench**, a diagnostic conversation benchmark designed to assess the relevance and quality of model-generated clinical reasoning. To further address the interpretability and alignment challenges of AI-assisted diagnosis, we develop a modular and medically grounded framework called **VAIDYA** that mirrors a physician’s stepwise diagnostic reasoning. This structured approach improves transparency and yields substantial performance gains over base LLMs. Our work takes a critical step toward aligning AI systems with real-world clinical practices by combining dynamic interaction, interpretability, and clinical validation.

VAIDYA: Validated Agents for Intelligent Diagnosis and Yielded Analysis

Self-Anchoring Calibration Drift} (SACD), a tendency for large language models (LLMs) to show systematic changes in expressed confidence when building iteratively on their own prior outputs across multi-turn conversations. Through a controlled three-condition study comparing Claude Sonnet 4.6, Gemini 3.1 Pro, and GPT-5.2 across factual, technical, and open-ended domains, we find that SACD is real but multiform: models exhibit distinct self-anchoring signatures ranging from active confidence suppression to calibration improvement suppression, with effects concentrated in open-ended domains. These findings challenge the adequacy of single-turn calibration evaluation for characterizing LLM reliability in realistic multi-turn deployment contexts. Code and data are available at https://github.com/hvardhan878/calibration-drift

Self-Anchoring Calibration Drift in Large Language Models: How Multi-Turn Conversations Reshape Model Confidence

Representing continuous time is a critical and under-explored challenge in modeling temporal event sequences with large language models (LLMs). Various strategies like byte-level representations or calendar tokens have been proposed. However, the optimal approach remains unclear, especially given the diverse statistical distributions of real-world event data, which range from smooth log-normal to discrete, spiky patterns. This paper presents a systematic empirical study of temporal tokenization for modeling event sequences with LLMs, comparing distinct encoding strategies: naive numeric strings, high-precision byte-level representations, human-semantic calendar tokens, classic uniform binning, and adaptive residual scalar quantization. We evaluate these strategies by fine-tuning LLMs on real-world datasets that exemplify these diverse distributions. Our analysis reveals that no single strategy is universally superior; instead, prediction performance depends heavily on aligning the tokenizer with the data's statistical properties, highlighting temporal tokenization as a critical yet often overlooked design dimension in LLM-based event modeling.

Temporal Tokenization Strategies for Event Sequence Modeling with Large Language Models

We present a large-scale human evaluation benchmark for assessing cultural localisation in machine translation produced by state-of-the-art multilingual large language models (LLMs). Existing MT benchmarks emphasise token-level and grammatical accuracy, but often overlook pragmatic and culturally grounded competencies required for real-world localisation. Building on a pilot study of 87 translations across 20 languages, we evaluate 7 multilingual LLMs across 15 target languages with 5 native-speaker raters per language. Raters scored both full-text translations and segment-level instances of culturally nuanced language (idioms, puns, holidays, and culturally embedded concepts) on an ordinal 0–3 quality scale; segment ratings additionally included an NA option for untranslated segments. Across full-text evaluations, mean overall quality is modest (1.68/3): GPT-5 (2.10/3), Claude Sonnet 3.7 (1.97/3), and Mistral Medium 3.1 (1.84/3) form the strongest tier with fewer catastrophic failures. Segment-level results show sharp category effects: holidays (2.20/3) and cultural concepts (2.19/3) translate substantially better than idioms (1.65/3) and puns (1.45/3), and idioms are most likely to be left untranslated. These findings demonstrate a persistent gap between grammatical adequacy and cultural resonance. To our knowledge, this is the first multilingual, human-annotated benchmark focused explicitly on cultural nuance in translation and localisation, highlighting the need for culturally informed training data, improved cross-lingual pragmatics, and evaluation paradigms that better reflect real-world communicative competence.

“Be My Cheese?”: Cultural Nuance Benchmarking for Machine Translation in Multilingual LLMs

The field of natural language generation has accumulated a rich ecosystem of automatic evaluation metrics, yet it lacks a coherent theory of what those metrics are actually measuring. Drawing on measurement theory from the quantitative social sciences, this paper argues that current NLG evaluation practices suffer from a fundamental construct validity problem: metrics are treated as proxies for output quality without explicit specification of the underlying constructs they are meant to operationalize. We examine four dominant evaluation paradigms (reference-based metrics, embedding-based metrics, LLM-as-judge, and human evaluation) and demonstrate that each conflates construct definition with operationalization. Building on a long psychometric tradition reaching back to Cronbach and Meehl (1955) and on recent NLP work that has begun to apply this tradition to bias measurement, dialogue evaluation, and benchmark design, we propose that the field adopt a measurement modeling perspective for NLG evaluation. We borrow the concepts of construct validity, reliability, and consequential validity as a foundation for more principled evaluation, and we outline a preliminary taxonomy of NLG quality constructs as a starting point for this work.

Position: What Are We Measuring? Rethinking Evaluation in Natural Language Generation

Post-training methods such as supervised fine-tuning and preference optimization are widely used to align large language models, yet how their benefits distribute across architectural components and transfer across tasks and prompts remains unclear. In this work, we analyze component-level transfer in a Mixture-of-Experts language model by selectively replacing routers, attention modules, and expert networks between two post-trained Mixture of Experts models trained with different post-training recipes and dataset mixtures. Starting from a SFT+DPO checkpoint, we systematically replace its components (routers, attention, experts) with those from a Tulu3 checkpoint and evaluate the impact of each replacement and their combinations on mathematical and scientific reasoning and a general-purpose classification task under zero-shot, few-shot and Chain of Thought prompting. We find strong component-specific specialization: expert networks account for most gains on mathematical and scientific reasoning, while attention mechanisms consistently outweigh expert transfer on general tasks and router transfer alone provides minimal benefit or harms performance. Prompting strategy further modulates these effects, with expert transfer degrading zero-shot science performance but improving few-shot reasoning. Strategically combining components from different model versions can in some cases match or exceed the performance of the best available model, motivating principled techniques for composing post-trained models into task- and prompt-specific systems without additional training.

Downloads

Next from ACL 2026

Theory-Explicit Prompting for MIND Self-States: Hierarchical LLMs and Dynamic Signature Extraction in Mental Health Timelines

Similar lecture

Overview of the CLPsych 2025 Shared Task: Capturing Mental Health Dynamics from Social Media Timelines

Stay up to date with the latest Underline news!

PRODUCT

COMPANY

RESOURCES