pith. sign in

arxiv: 2606.11875 · v1 · pith:OYO5P7GAnew · submitted 2026-06-10 · 💻 cs.CL · cs.SD

I Understand How You Feel: Enhancing Deeper Emotional Support Through Multilingual Emotional Validation in Dialogue System

Zi Haur Pang , Yahui Fu , Koji Inoue , Tatsuya Kawahara This is my paper

Pith reviewed 2026-06-27 09:43 UTC · model grok-4.3

classification 💻 cs.CL cs.SD
keywords emotional validationdialogue systemsmultilingual corporaemotion timing detectiongated fusioncross-modal attentionlarge language models
0
0 comments X

The pith

MEGUMI fuses semantic and emotion encoders to improve when dialogue systems should validate user feelings in English and Japanese.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper decomposes emotional validation into three parts: spotting when a validating response is needed, deciding the right moment to deliver it, and generating the response itself. It supplies two new resources, a large hybrid-annotated English-Japanese corpus and a spoken test set, so that each part can be studied separately. MEGUMI is introduced as a gated fusion architecture that keeps a frozen multilingual semantic model while adding language-specific emotion encoders and cross-modal attention, and this architecture records higher accuracy on timing detection than prior methods on both datasets. The same work benchmarks current large language models and finds they can produce varied validating replies yet still fall short on grasping the underlying emotion. A reader would care because explicit emotional validation carries documented therapeutic value and most existing dialogue systems lack it.

Core claim

By releasing M-EDESConv and M-TESC and training MEGUMI to combine frozen XLM-RoBERTa representations with language-specific emotion encoders through cross-modal attention and gated fusion, the authors establish superior objective and subjective results on the task of detecting the correct timing for emotional validation responses in multilingual dialogue.

What carries the argument

MEGUMI (Multilingual Emotion-aware Gated Unit for Mutual Integration), which fuses frozen XLM-RoBERTa semantics with language-specific emotion encoders via cross-modal attention and gated fusion to decide validation timing.

If this is right

  • Dialogue systems equipped with MEGUMI-style timing detection can deliver validating replies at moments that better match user expectations.
  • The M-EDESConv corpus directly supports separate study of validating response identification, timing detection, and response generation.
  • Multilingual training data allows the same validation timing model to operate across English and Japanese without separate monolingual systems.
  • Current large language models already produce contextually appropriate and diverse validating responses once the timing cue is supplied.
  • Emotional understanding remains the main bottleneck for large language models on this task.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The gated-fusion design could be tested on additional language pairs to check whether the same architecture scales beyond English and Japanese.
  • If timing detection improves, downstream user studies could measure whether users report higher perceived support or lower distress after interacting with the system.
  • The hybrid annotation pipeline may need explicit checks for cultural differences in what counts as validation before the datasets are used to train models for new language communities.

Load-bearing premise

The hybrid manual-automatic labels in M-EDESConv faithfully mark genuine emotional validation without large amounts of noise or cultural bias.

What would settle it

A blinded human rating study on held-out dialogues in which raters mark the moments they would expect validation; if MEGUMI's predicted timings receive lower agreement or preference scores than strong baselines, the claimed superiority on timing detection is falsified.

Figures

Figures reproduced from arXiv: 2606.11875 by Koji Inoue, Tatsuya Kawahara, Yahui Fu, Zi Haur Pang.

Figure 1
Figure 1. Figure 1: Emotional validation can be decomposed into three subtasks: (1) Validating Response Identification, [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Overall proposed architecture in this study. We proposed a Multilingual Emotion-aware Gated Unit for [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
read the original abstract

Emotional validation - explicitly acknowledging that a user's feelings make sense - has proven therapeutic value but has received little computational attention. Emotional validation in dialogue systems can be decomposed into (i) validating response identification, (ii) validation timing detection, and (iii) validating response generation. To support research on all three subtasks, we release M-EDESConv, a 120k English-Japanese multilingual corpus created through hybrid manual and automatic annotation, and M-TESC, a multilingual spoken-dialogue test set. For timing detection, we propose MEGUMI, a Multilingual Emotion-aware Gated Unit for Mutual Integration, that fuses frozen XLM-RoBERTa semantics with language-specific emotion encoders via cross-modal attention and gated fusion. MEGUMI shows superior performance on both the M-EDESConv and M-TESC datasets, both objectively and subjectively. Finally, our EmoValidBench benchmarks of GPT-4.1 Nano and Llama-3.1 8B indicate that current LLMs generate contextually similar and diverse validating responses, but emotional understanding remains a major area for improvement. Project page: https://github.com/zihaurpang/Multilingual-Emotional-Validation

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper claims that emotional validation in dialogue systems decomposes into validating response identification, timing detection, and response generation. It releases M-EDESConv (120k English-Japanese utterances via hybrid manual-automatic annotation) and M-TESC (multilingual spoken test set) to support research on these subtasks, proposes MEGUMI (a model fusing frozen XLM-RoBERTa semantics with language-specific emotion encoders via cross-modal attention and gated fusion), reports that MEGUMI achieves superior objective and subjective performance on both datasets, and shows via EmoValidBench that current LLMs produce contextually similar and diverse validating responses but still lack strong emotional understanding.

Significance. If the results hold after verification of label quality, the work supplies the first large-scale multilingual resources explicitly targeting emotional validation and a model architecture tailored to timing detection across languages. Dataset release and the explicit three-subtask framing are concrete strengths that could enable follow-on work in therapeutic dialogue systems.

major comments (2)
  1. [Corpus construction (abstract and §3)] Corpus construction (abstract and §3): The hybrid manual-automatic annotation process for the 120k M-EDESConv labels is described without inter-annotator agreement figures, without any held-out human validation of the automatic component, and without cross-lingual consistency checks. Because every reported MEGUMI performance number (both on M-EDESConv and on M-TESC) depends on these labels faithfully encoding emotional validation, the absence of these diagnostics is load-bearing for the central superiority claim.
  2. [Experiments (§5)] Experiments (§5): No ablation isolating the contribution of the gated fusion versus the cross-modal attention, no error bars or statistical significance tests on the reported objective metrics, and no breakdown of performance by language or by validation timing difficulty. These omissions prevent assessment of whether the claimed gains are robust or driven by particular subsets of the (unvalidated) labels.
minor comments (2)
  1. [Figure 2] Figure 2 (model diagram): The gated fusion block is shown schematically but the exact equations for the gate computation and the cross-modal attention are not written out; adding them would improve reproducibility.
  2. [Related work] Related work: The positioning relative to prior emotional support dialogue datasets (e.g., ESConv) is brief; a short table comparing label granularity and language coverage would clarify the novelty of M-EDESConv.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback. We address each major comment below and will revise the manuscript accordingly to strengthen the claims regarding label quality and experimental robustness.

read point-by-point responses

  1. Referee: [Corpus construction (abstract and §3)] Corpus construction (abstract and §3): The hybrid manual-automatic annotation process for the 120k M-EDESConv labels is described without inter-annotator agreement figures, without any held-out human validation of the automatic component, and without cross-lingual consistency checks. Because every reported MEGUMI performance number (both on M-EDESConv and on M-TESC) depends on these labels faithfully encoding emotional validation, the absence of these diagnostics is load-bearing for the central superiority claim.

    Authors: We agree that these diagnostics are necessary to support the reliability of the labels. In the revised manuscript we will report inter-annotator agreement for the manual portion of the annotation, accuracy results from a held-out human validation set for the automatic component, and cross-lingual consistency metrics between English and Japanese. These additions will directly address the concern that label quality underpins the reported performance numbers. revision: yes

  2. Referee: [Experiments (§5)] Experiments (§5): No ablation isolating the contribution of the gated fusion versus the cross-modal attention, no error bars or statistical significance tests on the reported objective metrics, and no breakdown of performance by language or by validation timing difficulty. These omissions prevent assessment of whether the claimed gains are robust or driven by particular subsets of the (unvalidated) labels.

    Authors: We concur that the current experimental section would benefit from these analyses. The revised version will include an ablation study isolating the gated fusion and cross-modal attention components, error bars with statistical significance tests on all objective metrics, and performance breakdowns by language and by validation timing difficulty. These changes will allow readers to evaluate the robustness of the gains. revision: yes

Circularity Check

0 steps flagged

No circularity; empirical model evaluation on released datasets

full rationale

The paper contains no equations, derivations, or parameter-fitting steps that could reduce to self-definition or fitted-input-as-prediction. MEGUMI is presented as a novel architecture fusing existing encoders, with superiority claims resting solely on held-out test performance (objective metrics plus subjective evaluation) on the newly released M-EDESConv and M-TESC corpora. The hybrid annotation process is a data-construction detail, not a load-bearing derivation or self-citation chain. No self-citations are invoked to justify uniqueness or ansatzes, and no renaming of known results occurs. This is a standard empirical NLP contribution whose central claims are externally falsifiable via the released data and code.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Abstract-only view yields limited visibility into modeling assumptions; the three-way decomposition of emotional validation is treated as given without further justification.

axioms (1)
  • domain assumption Emotional validation in dialogue systems decomposes cleanly into validating response identification, validation timing detection, and validating response generation.
    Stated in the abstract as the basis for the three subtasks and dataset design.

pith-pipeline@v0.9.1-grok · 5751 in / 1147 out tokens · 14813 ms · 2026-06-27T09:43:37.438523+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

100 extracted references · 16 canonical work pages · 7 internal anchors

  1. [1]

    MIME : MIM icking Emotions for Empathetic Response Generation

    Majumder, Navonil and Hong, Pengfei and Peng, Shanshan and Lu, Jiankun and Ghosal, Deepanway and Gelbukh, Alexander and Mihalcea, Rada and Poria, Soujanya. MIME : MIM icking Emotions for Empathetic Response Generation. Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP). 2020

    2020

  2. [2]

    Proceedings of the AAAI Conference on Artificial Intelligence , volume=

    Cem: Commonsense-aware empathetic response generation , author=. Proceedings of the AAAI Conference on Artificial Intelligence , volume=

  3. [3]

    Towards Empathetic Open-domain Conversation Models: A New Benchmark and Dataset

    Rashkin, Hannah and Smith, Eric Michael and Li, Margaret and Boureau, Y-Lan. Towards Empathetic Open-domain Conversation Models: A New Benchmark and Dataset. Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics. 2019

    2019

  4. [4]

    Towards Emotional Support Dialog Systems

    Liu, Siyang and Zheng, Chujie and Demasi, Orianna and Sabour, Sahand and Li, Yu and Yu, Zhou and Jiang, Yong and Huang, Minlie. Towards Emotional Support Dialog Systems. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021

    2021

  5. [5]

    arXiv preprint arXiv:2311.15732 , year=

    GPT4Vis: what can GPT-4 do for zero-shot visual recognition? , author=. arXiv preprint arXiv:2311.15732 , year=

    work page arXiv

  6. [6]

    arXiv preprint arXiv:2402.11409 , year=

    Multi-dimensional Evaluation of Empathetic Dialog Responses , author=. arXiv preprint arXiv:2402.11409 , year=

    work page arXiv

  7. [7]

    G o E motions: A Dataset of Fine-Grained Emotions

    Demszky, Dorottya and Movshovitz-Attias, Dana and Ko, Jeongwoo and Cowen, Alan and Nemade, Gaurav and Ravi, Sujith. G o E motions: A Dataset of Fine-Grained Emotions. Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. 2020

    2020

  8. [8]

    Gated Multimodal Units for Information Fusion

    Gated multimodal units for information fusion , author=. arXiv preprint arXiv:1702.01992 , year=

    work page internal anchor Pith review Pith/arXiv arXiv

  9. [9]

    Proceedings of the conference

    Multimodal transformer for unaligned multimodal language sequences , author=. Proceedings of the conference. Association for computational linguistics. Meeting , volume=

  10. [10]

    arXiv preprint arXiv:2505.00013 , year=

    Performance Evaluation of Emotion Classification in Japanese Using RoBERTa and DeBERTa , author=. arXiv preprint arXiv:2505.00013 , year=

    work page arXiv

  11. [11]

    Biomedical informatics insights , volume=

    A hybrid model for automatic emotion recognition in suicide notes , author=. Biomedical informatics insights , volume=. 2012 , publisher=

    2012

  12. [12]

    2022 ieee spoken language technology workshop (slt) , pages=

    Empirical analysis of training strategies of transformer-based japanese chit-chat systems , author=. 2022 ieee spoken language technology workshop (slt) , pages=. 2023 , organization=

    2022

  13. [13]

    Educational technology research and development , pages=

    Pedagogical AI conversational agents in higher education: a conceptual framework and survey of the state of the art , author=. Educational technology research and development , pages=. 2025 , publisher=

    2025

  14. [14]

    Empathy and Business Transformation , pages=

    Human or nonhuman agent?: Experiences of empathy in a digital customer tourism journey , author=. Empathy and Business Transformation , pages=. 2022 , publisher=

    2022

  15. [15]

    In Proceedings of 2023 IEEE International Conference on Robot & Human Interactive Communication (RO-MAN) , year=

    Developing social robots with empathetic non-verbal cues using large language models , author=. In Proceedings of 2023 IEEE International Conference on Robot & Human Interactive Communication (RO-MAN) , year=

    2023

  16. [16]

    Findings of the association for computational linguistics: EMNLP 2021 , pages=

    Improving empathetic response generation by recognizing emotion cause in conversations , author=. Findings of the association for computational linguistics: EMNLP 2021 , pages=

    2021

  17. [17]

    ICASSP 2024-2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) , pages=

    Pecer: Empathetic response generation via dynamic personality extraction and contextual emotional reasoning , author=. ICASSP 2024-2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) , pages=. 2024 , organization=

    2024

  18. [18]

    Towards Persona-Based Empathetic Conversational Models

    Zhong, Peixiang and Zhang, Chen and Wang, Hao and Liu, Yong and Miao, Chunyan. Towards Persona-Based Empathetic Conversational Models. Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP). 2020

    2020

  19. [19]

    Processes , volume=

    Development of an empathy-centric counseling chatbot system capable of sentimental dialogue analysis , author=. Processes , volume=. 2022 , publisher=

    2022

  20. [20]

    Immersive Learning Research-Practitioner , pages=

    Empathetic chatbot: Enhancing medical education with artificial intelligence , author=. Immersive Learning Research-Practitioner , pages=

  21. [21]

    Journal of the Academy of Marketing Science , volume=

    Artificial empathy in marketing interactions: Bridging the human-AI gap in affective and social customer experience , author=. Journal of the Academy of Marketing Science , volume=. 2022 , publisher=

    2022

  22. [22]

    arXiv preprint arXiv:2408.11599 , year=

    Cause-Aware Empathetic Response Generation via Chain-of-Thought Fine-Tuning , author=. arXiv preprint arXiv:2408.11599 , year=

    work page arXiv

  23. [23]

    Computers in Human Behavior , volume=

    Nonverbal behaviors perceived as most empathic in a simulated medical context , author=. Computers in Human Behavior , volume=. 2024 , publisher=

    2024

  24. [24]

    Findings of the Association for Computational Linguistics: NAACL 2022 , pages=

    Empathetic persuasion: reinforcing empathy and persuasiveness in dialogue systems , author=. Findings of the Association for Computational Linguistics: NAACL 2022 , pages=

    2022

  25. [25]

    Proceedings of the 24th Annual Meeting of the Special Interest Group on Discourse and Dialogue , pages=

    Empathetic response generation for distress support , author=. Proceedings of the 24th Annual Meeting of the Special Interest Group on Discourse and Dialogue , pages=

  26. [26]

    ACM Transactions on Human-Robot Interaction , volume=

    The effects of healthcare robot empathy statements and head nodding on trust and satisfaction: a video study , author=. ACM Transactions on Human-Robot Interaction , volume=. 2023 , publisher=

    2023

  27. [27]

    Borderline Personality Disorder and Emotion Dysregulation , volume=

    The who and what of validation: an experimental examination of validation and invalidation of specific emotions and the moderating effect of emotion dysregulation , author=. Borderline Personality Disorder and Emotion Dysregulation , volume=. 2022 , publisher=

    2022

  28. [28]

    Patient Experience Journal , volume=

    Is Timing Everything?: The Role of Time on the Relationship between Patient-Centered Communication and Provider Empathy , author=. Patient Experience Journal , volume=

  29. [29]

    , author=

    Validation and psychotherapy. , author=. 1997 , journal=

    1997

  30. [30]

    Exploring emotional validation in cross-cultural management: a case study , author=

  31. [31]

    Scandinavian Journal of Pain , volume=

    I see you're in pain--The effects of partner validation on emotions in people with chronic pain , author=. Scandinavian Journal of Pain , volume=. 2015 , publisher=

    2015

  32. [32]

    Computational linguistics , volume=

    Dialogue act modeling for automatic tagging and recognition of conversational speech , author=. Computational linguistics , volume=. 2000 , publisher=

    2000

  33. [33]

    Linguistics Vanguard , volume=

    Could this be next for corpus linguistics? Methods of semi-automatic data annotation with contextualized word embeddings , author=. Linguistics Vanguard , volume=. 2024 , publisher=

    2024

  34. [34]

    The 37th Annual Conference of the Japanese Society for Artificial Intelligence , pages=

    Prediction of Validating Response from Emotional Storytelling Corpus , author=. The 37th Annual Conference of the Japanese Society for Artificial Intelligence , pages=

  35. [35]

    Proceedings of the Workshop on Computational Modeling of People's Opinions, Personality, and Emotions in Social Media (PEOPLES) , pages=

    Innovative semi-automatic methodology to annotate emotional corpora , author=. Proceedings of the Workshop on Computational Modeling of People's Opinions, Personality, and Emotions in Social Media (PEOPLES) , pages=

  36. [36]

    E mp H i: Generating Empathetic Responses with Human-like Intents

    Chen, Mao Yan and Li, Siheng and Yang, Yujiu. E mp H i: Generating Empathetic Responses with Human-like Intents. Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2022

    2022

  37. [37]

    A Taxonomy of Empathetic Response Intents in Human Social Conversations

    Welivita, Anuradha and Pu, Pearl. A Taxonomy of Empathetic Response Intents in Human Social Conversations. Proceedings of the 28th International Conference on Computational Linguistics. 2020

    2020

  38. [38]

    Signals , volume=

    Dialogue act classification via transfer learning for automated labeling of interviewee responses in virtual reality job interview training platforms for autistic individuals , author=. Signals , volume=. 2023 , publisher=

    2023

  39. [39]

    Archives of general psychiatry , volume=

    Trauma and posttraumatic stress disorder in the community: the 1996 Detroit Area Survey of Trauma , author=. Archives of general psychiatry , volume=. 1998 , publisher=

    1996

  40. [40]

    arXiv preprint arXiv:2412.09867 , year=

    Human-Like Embodied AI Interviewer: Employing Android ERICA in Real International Conference , author=. arXiv preprint arXiv:2412.09867 , year=

    work page arXiv

  41. [41]

    Lee, Akinobu , license =

  42. [42]

    9th International Workshop on Spoken Dialogue System Technology , pages=

    Spoken dialogue system for a human-like conversational robot ERICA , author=. 9th International Workshop on Spoken Dialogue System Technology , pages=. 2019 , organization=

    2019

  43. [43]

    Proceedings of the 18th Annual SIGdial Meeting on Discourse and Dialogue , pages=

    Attentive listening system with backchanneling, response generation and flexible turn-taking , author=. Proceedings of the 18th Annual SIGdial Meeting on Discourse and Dialogue , pages=

  44. [44]

    International Journal of Social Robotics , volume=

    Enhancing emotional support: The effect of a robotic object on human--human support quality , author=. International Journal of Social Robotics , volume=. 2022 , publisher=

    2022

  45. [45]

    arXiv preprint arXiv:2503.13625 , year=

    Does the Appearance of Autonomous Conversational Robots Affect User Spoken Behaviors in Real-World Conference Interactions? , author=. arXiv preprint arXiv:2503.13625 , year=

    work page arXiv

  46. [46]

    , author=

    The effect of therapist use of validation strategies on change in client emotion in individual dbt treatment sessions. , author=. Personality Disorders: Theory, Research, and Treatment , volume=. 2018 , publisher=

    2018

  47. [47]

    Merrill-Palmer Quarterly , volume=

    The role of maternal emotional validation and invalidation on children's emotional awareness , author=. Merrill-Palmer Quarterly , volume=. 2016 , publisher=

    2016

  48. [48]

    Administration and Policy in Mental Health and Mental Health Services Research , pages=

    Youths' perceived impact of invalidation and validation on their mental health treatment journeys , author=. Administration and Policy in Mental Health and Mental Health Services Research , pages=. 2022 , publisher=

    2022

  49. [49]

    14th International Workshop on Spoken Dialogue System Technology , year=

    Acknowledgment of emotional states: Generating validating responses for empathetic dialogue , author=. 14th International Workshop on Spoken Dialogue System Technology , year=

  50. [50]

    arXiv preprint arXiv:2404.01657 , year=

    Release of pre-trained models for the Japanese language , author=. arXiv preprint arXiv:2404.01657 , year=

    work page arXiv

  51. [51]

    My child will actually say ‘I am upset'… Before all they would do was scream

    “My child will actually say ‘I am upset'… Before all they would do was scream”: Teaching parents emotion validation in a social care setting , author=. Child: Care, Health and Development , volume=. 2020 , publisher=

    2020

  52. [52]

    Advanced Robotics , volume=

    Dual variational generative model and auxiliary retrieval for empathetic response generation by conversational robot , author=. Advanced Robotics , volume=. 2023 , publisher=

    2023

  53. [53]

    Proceedings of the AAAI conference on artificial intelligence , volume=

    A hierarchical latent variable encoder-decoder model for generating dialogues , author=. Proceedings of the AAAI conference on artificial intelligence , volume=

  54. [54]

    arXiv preprint arXiv:1905.05471 , year=

    Improving neural conversational models with entropy-based data filtering , author=. arXiv preprint arXiv:1905.05471 , year=

    work page arXiv 1905

  55. [55]

    How NOT To Evaluate Your Dialogue System: An Empirical Study of Unsupervised Evaluation Metrics for Dialogue Response Generation

    How not to evaluate your dialogue system: An empirical study of unsupervised evaluation metrics for dialogue response generation , author=. arXiv preprint arXiv:1603.08023 , year=

    work page internal anchor Pith review Pith/arXiv arXiv

  56. [56]

    A Diversity-Promoting Objective Function for Neural Conversation Models

    Li, Jiwei and Galley, Michel and Brockett, Chris and Gao, Jianfeng and Dolan, Bill. A Diversity-Promoting Objective Function for Neural Conversation Models. Proceedings of the 2016 Conference of the North A merican Chapter of the Association for Computational Linguistics: Human Language Technologies. 2016

    2016

  57. [57]

    Proceedings of the 40th annual meeting of the Association for Computational Linguistics , pages=

    Bleu: a method for automatic evaluation of machine translation , author=. Proceedings of the 40th annual meeting of the Association for Computational Linguistics , pages=

  58. [58]

    International Conference on Learning Representations , year=

    BERTScore: Evaluating Text Generation with BERT , author=. International Conference on Learning Representations , year=

  59. [59]

    Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks

    Sentence-bert: Sentence embeddings using siamese bert-networks , author=. arXiv preprint arXiv:1908.10084 , year=

    work page internal anchor Pith review Pith/arXiv arXiv 1908

  60. [60]

    Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) , pages=

    A Close Look into the Calibration of Pre-trained Language Models , author=. Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) , pages=

  61. [61]

    JAMIA open , volume=

    Decoding disparities: evaluating automatic speech recognition system performance in transcribing Black and White patient verbal communication with nurses in home healthcare , author=. JAMIA open , volume=. 2024 , publisher=

    2024

  62. [62]

    Unsupervised Cross-lingual Representation Learning at Scale

    Conneau, Alexis and Khandelwal, Kartikay and Goyal, Naman and Chaudhary, Vishrav and Wenzek, Guillaume and Guzm \'a n, Francisco and Grave, Edouard and Ott, Myle and Zettlemoyer, Luke and Stoyanov, Veselin. Unsupervised Cross-lingual Representation Learning at Scale. Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. 2020

    2020

  63. [63]

    Advances in neural information processing systems , volume=

    Large language models are zero-shot reasoners , author=. Advances in neural information processing systems , volume=

  64. [64]

    Advances in neural information processing systems , volume=

    Chain-of-thought prompting elicits reasoning in large language models , author=. Advances in neural information processing systems , volume=

  65. [65]

    2021 24th Conference of the Oriental COCOSDA International Committee for the Co-ordination and Standardisation of Speech Databases and Assessment Techniques (O-COCOSDA) , pages=

    Design and Basic Analysis of the TUT Emotional Storytelling Corpus , author=. 2021 24th Conference of the Oriental COCOSDA International Committee for the Co-ordination and Standardisation of Speech Databases and Assessment Techniques (O-COCOSDA) , pages=. 2021 , organization=

    2021

  66. [66]

    American scientist , volume=

    The nature of emotions: Human emotions have deep evolutionary roots, a fact that may explain their complexity and provide tools for clinical practice , author=. American scientist , volume=. 2001 , publisher=

    2001

  67. [67]

    Proceedings of the 25th Annual Meeting of the Special Interest Group on Discourse and Dialogue , pages=

    StyEmp: Stylizing Empathetic Response Generation via Multi-Grained Prefix Encoder and Personality Reinforcement , author=. Proceedings of the 25th Annual Meeting of the Special Interest Group on Discourse and Dialogue , pages=

  68. [68]

    29th COLING , pages=

    Does GPT-3 Generate Empathetic Dialogues? A Novel In-Context Example Selection Method and Automatic Evaluation Metric for Empathetic Dialogue Generation , author=. 29th COLING , pages=

  69. [69]

    RoBERTa: A Robustly Optimized BERT Pretraining Approach

    Roberta: A robustly optimized bert pretraining approach , author=. arXiv preprint arXiv:1907.11692 , year=

    work page internal anchor Pith review Pith/arXiv arXiv 1907

  70. [70]

    No Language Left Behind: Scaling Human-Centered Machine Translation

    No language left behind: Scaling human-centered machine translation , author=. arXiv preprint arXiv:2207.04672 , year=

    work page internal anchor Pith review Pith/arXiv arXiv

  71. [71]

    International journal of human-computer studies , volume=

    The influence of empathy in human--robot relations , author=. International journal of human-computer studies , volume=. 2013 , publisher=

    2013

  72. [72]

    Journal of medical Internet research , volume=

    Towards an artificially empathic conversational agent for mental health applications: system design and user perceptions , author=. Journal of medical Internet research , volume=. 2018 , publisher=

    2018

  73. [73]

    Proceedings of the 21st ACM international conference on intelligent virtual agents , pages=

    Enhancing conversational agents with empathic abilities , author=. Proceedings of the 21st ACM international conference on intelligent virtual agents , pages=

  74. [74]

    Proceedings of the 24th Annual Meeting of the Special Interest Group on Discourse and Dialogue , pages=

    Reasoning before Responding: Integrating Commonsense-based Causality Explanation for Empathetic Response Generation , author=. Proceedings of the 24th Annual Meeting of the Special Interest Group on Discourse and Dialogue , pages=

  75. [75]

    Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP) , pages=

    A Computational Approach to Understanding Empathy Expressed in Text-Based Mental Health Support , author=. Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP) , pages=

    2020

  76. [76]

    M o EL : Mixture of Empathetic Listeners

    Lin, Zhaojiang and Madotto, Andrea and Shin, Jamin and Xu, Peng and Fung, Pascale. M o EL : Mixture of Empathetic Listeners. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). 2019

    2019

  77. [77]

    Proceedings of the AAAI conference on artificial intelligence , volume=

    Knowledge bridging for empathetic dialogue generation , author=. Proceedings of the AAAI conference on artificial intelligence , volume=

  78. [78]

    Can Large Language Models be Good Emotional Supporter? Mitigating Preference Bias on Emotional Support Conversation

    Kang, Dongjin and Kim, Sunghwan and Kwon, Taeyoon and Moon, Seungjun and Cho, Hyunsouk and Yu, Youngjae and Lee, Dongha and Yeo, Jinyoung. Can Large Language Models be Good Emotional Supporter? Mitigating Preference Bias on Emotional Support Conversation. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Lo...

    2024

  79. [79]

    ESC o T : Towards Interpretable Emotional Support Dialogue Systems

    Zhang, Tenggan and Zhang, Xinjie and Zhao, Jinming and Zhou, Li and Jin, Qin. ESC o T : Towards Interpretable Emotional Support Dialogue Systems. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024

    2024

  80. [80]

    Frontiers in Psychiatry , volume=

    Enhancing the conversational agent with an emotional support system for mental health digital therapeutics , author=. Frontiers in Psychiatry , volume=. 2023 , publisher=

    2023

Showing first 80 references.