The YNU-HPCC team participated in Subtask A of SemEval‑2025 Task 11, "Bridging the Gap in Text‑Based Emotion," using a RoBERTa model with a single prediction head to process one emotion at a time. Their system achieved an official ranking score of 0.44 across all languages after translating the dataset into English with Google Translate. Analysis showed that a single head outperformed six simultaneous heads and that training on the uniformly translated English data improved results.
arXiv:2403.17612v3 Announce Type: replace Abstract: Labeling corpora constitutes a bottleneck to create models for new tasks or domains. Large language models mitigate the issue with automatic corpus...
arXiv:2608.29035v1 Announce Type: new Abstract: Emotion recognition in conversations is increasingly tackled with language models, but these models can be unstable and expensive to fine-tune or to pr...
arXiv:2601. 03888v4 Announce Type: replace-cross Abstract: In prior work, we introduced IndexTTS 2, a zero-shot neural text-to-speech foundation model comprising two core components: a transformer-based Text-to-Semantic (T2S) module and a non-autoregressive Semantic-to-Mel (S2M) module, which together enable faithful emotion replication and establish the first autoregressive duration-controllable generative paradigm.
arXiv:2606. 27717v1 Announce Type: cross Abstract: Prosodic emphasis varies across languages, emotions, and speaking styles, yet existing emphasis detection models are largely trained and evaluated on monolingual neutral read speech.
arXiv:2608.29613v1 Announce Type: cross Abstract: Function vectors (FVs) have recently emerged as a promising mechanism for steering the behavior of large language models (LLMs) by injecting task-spe...
arXiv:2309.15670v3 Announce Type: replace Abstract: In recent years, Sentiment Analysis (SA) and Emotion Recognition (ER) have been increasingly popular in the Bangla language, which is the seventh m...
The paper introduces a discriminative adaptation for SpeechLLMs that reads the hidden state of the final prompt token via a simple classification head, enabling emotion recognition in a single forward pass without altering the backbone. This approach replaces the generative decoder, which can produce out‑of‑set labels and favor frequent classes, with a controlled comparison between generative and discriminative inference. Experiments on IEMOCAP show improved Macro F1 scores, elimination of hallucinations, and larger gains on realistic ASR transcripts, while revealing that emotion directions encode indirect associations reflecting web‑scale text biases.
TalkFa introduces a unified benchmark for Farsi dialogue generation and understanding, comprising three datasets: WIKI‑FADIAL (4.2K Wikipedia‑grounded dialogues), DAILYDIALOG‑FA (6.6K dialogues with dialogue‑act and emotion annotations), and PLAYDIAL‑FA (2.1K theatrical dialogues with sentiment labels). All dialogues are curated through multi‑stage review by native speakers, ensuring high quality. Experiments show that LoRA fine‑tuning improves generation performance with less data, while specific models excel on classification tasks, and human evaluation confirms the benchmark’s reliability.
We present a Procrustes-conditioned Joint End-to-end Top-K Sparse Autoencoder (SAE) for extracting cross-seed universal features from independently trained BERT models. Cross-seed feature universality is a fundamental challenge in mechanistic interpretability: because dictionary learning is non-convex, independently trained networks learn misaligned feature spaces, so apparently identical features may differ by random initialization.
Understanding both expressed and evoked emotions is critical for multimodal large language models (MLLMs) to achieve comprehensive affect-aware interactions. However, existing benchmarks typically examine expressed and evoked emotions in isolation or are constrained to coarse-grained and incomplete affective characterizations.
arXiv:2609.36194v1 Announce Type: new Abstract: Extracted sentiment directions can vary across samples even when downstream sentiment classification remains accurate. To evaluate direction reproducib...
The paper introduces CHIARO, a 1,000-sentence benchmark for contrastive emotion inference grounded in appraisal theory, where each scenario elicits a positive emotion in one person and a negative emotion in another. The dataset covers ten emotion classes and is human‑annotated. Evaluation shows that the best large language model achieves 67.3 macro‑F1, below human agreement, while existing emotion classifiers perform near chance. When used as a training signal alongside an existing emotion corpus, models improve on CHIARO and on six of ten external emotion benchmarks, demonstrating its value as a complementary training resource.