The study examines how emotions are represented across layers of large language models (LLMs) by probing eight 1B–9B open‑weight models on three datasets (Twitter, Reddit, autobiographical narratives). It finds that the optimal probing layer varies systematically with the dataset, moving from near‑input layers to deeper layers, and that targeted forward‑pass interventions on these layers degrade performance more than random interventions. Additionally, the selected layers transfer across datasets and emotion categories, and early‑exit representations from these layers outperform full‑depth exits by an average of 6.9 percentage points.
By Tian Fang, Ga\"el Guibon, Davide Buscaldi
arXiv:2609.05806v1 Announce Type: new
Abstract: Emotion Recognition in Conversations (ERC) aims to identify speakers' emotions in multi-turn dialogue. Accurate emotion recognition can support a wide...
By Amir Ben Khalifa, Fanny Bezancon, Amine Trabelsi, Bessam Abdulrazak
DiaRelay introduces a lightweight adapter that lets large language models maintain a constant‑size dialogue‑level memory for emotion recognition in conversation. It builds on LoRA by adding a Selective Relay Memory Transition that aggregates useful historical evidence into a bounded memory, and a Dual‑axis Relay Memory Read that uses this memory to modulate low‑rank feature transformations. Experiments show DiaRelay achieves state‑of‑the‑art weighted F1 and accuracy on MELD with only 7.1 M additional trainable parameters, while also performing competitively on IEMOCAP.
By Zihao Zhou, Bin Yang, Jinghui Qin, Kebing Jin
arXiv:2609.36563v1 Announce Type: new
Abstract: Visual emotion recognition commonly assumes that all evidence required for prediction is contained in the observed image or video. Yet the same visible...
By Yihao Qian, Runhao Zeng, Sicheng Zhao, Feng Liang, Hongmin Cai, Mingkui Tan
arXiv:2609.39453v1 Announce Type: cross
Abstract: Speech emotion recognition (SER) is the task of assigning emotion labels to utterances. Early systems relied on acoustic features, whereas recent app...
By Hezhao Zhang, Thomas Hain
arXiv:2604. 19139v3 Announce Type: replace-cross Abstract: As Large Language Models (LLMs) continue to evolve through alignment techniques such as Reinforcement Learning from Human Feedback (RLHF) and Constitutional AI, a growing and increasingly conspicuous phenomenon has emerged: the proliferation of verbal tics--repetitive, formulaic linguistic patterns that pervade model outputs.
By Shuai Wu, Xue Li, Yanna Feng, Yufang Li, Zhijun Wang, Ran Wang
The study examines how the geometric placement of synthetic data affects discourse‑pragmatic function classification. Using 410 annotated instances of the word "look" from the British National Corpus, synthetic examples were generated with Llama 3.1 and grouped by cosine distance from real data in RoBERTa space. Six training conditions were compared, showing that examples close to real data (NEAR) yield the largest macro‑F gain, while a distance‑balanced mix gives the highest accuracy, yet none improve AUC.
By Sara Sorahi, Kevin Tang, Reza Kazemian
arXiv:2606. 00851v1 Announce Type: cross Abstract: Empathetic spoken dialogue systems must infer a user's emotional state to respond appropriately, yet everyday speech often carries weak, neutral, or ambiguous affective cues.
By Sukru Samet Dindar, Riki Shimizu, Xilin Jiang, Nima Mesgarani
arXiv:2609.09924v1 Announce Type: new
Abstract: Emotion Recognition in Conversations (ERC) requires integrating heterogeneous textual, audio, and visual cues while accounting for conversational conte...
By Oriol Mar\'in, Roger Mar\'i, Gloria Haro, Rafael Redondo
VoiceLongMemEval (VLME) is a new benchmark that tests AI assistants on their ability to remember how users sounded by incorporating paralinguistic metadata—such as emotion labels, prosody descriptors, and voice events—into each conversational turn. The benchmark uses a three‑stage adversarial gate to ensure that models cannot succeed with transcript alone, revealing a significant affect gap: models gain 0.09 to 0.38 accuracy when provided with paralinguistic cues, and audio‑native models outperform standard ASR pipelines in extracting these signals. The dataset and code will be released upon acceptance.
By Ramit Pahwa, Parivesh Priye, Apoorva Beedu
BiCFlow-MER introduces a conditional-flow framework for audio-text multimodal emotion recognition, treating the task as generative evidence transport within a structured emotion space. It disentangles emotion-oriented evidence from speaker style and lexical content, creating a conflict-aware affective condition that guides bidirectional rectified flow to an explicit emotion-space endpoint. The model verifies candidate emotions via adaptive prototype-cloud scoring and backward class-to-condition consistency, achieving superior performance on IEMOCAP, MELD, and the zero-shot CASE benchmark.
By Yanbing Wang, Shenyue Wang, Chunyang Yu
arXiv:2606. 26987v1 Announce Type: cross Abstract: Recent work identified emotion vectors in Claude Sonnet 4.
By Sinie van der Ben, Rapha\"el Baur, Yannick Metz, Mennatallah El-Assady