GeoLaux: A Benchmark for Evaluating MLLMs' Geometry Performance on Long-Step Problems Requiring Auxiliary Lines

Geometry problem solving (GPS) requires models to master diagram comprehension, logical reasoning, knowledge application, numerical computation, and auxiliary line construction. This presents a significant challenge for Multimodal Large Language Models (MLLMs). However, existing benchmarks for evaluating MLLM geometry skills overlook auxiliary line construction and lack fine-grained process evaluation, making them insufficient for assessing MLLMs' long-step reasoning abilities. To bridge these gaps, we present the GeoLaux benchmark, comprising 2,186 geometry problems, incorporating both calculation and proving questions. Notably, the problems require an average of 6.51 reasoning steps, with a maximum of 24 steps, and 41.8% of them need auxiliary line construction. Building on the dataset, we design a novel five-dimensional evaluation strategy assessing answer correctness, process correctness, process quality, auxiliary line impact, and error causes. Extensive experiments on 13 leading MLLMs (including thinking models and non-thinking models) yield three pivotal findings: First, models exhibit substantial performance degradation in extended reasoning steps (nine models demonstrate over 50% performance drop). Second, compared to calculation problems, MLLMs tend to take shortcuts when solving proving problems. Third, models lack auxiliary line awareness, and enhancing this capability proves particularly beneficial for overall geometry reasoning improvement. These findings establish GeoLaux as both a benchmark for evaluating MLLMs' long-step geometric reasoning with auxiliary lines and a guide for capability advancement. Our dataset and code are included in supplementary materials and will be released.

BeMERC: Behavior-Aware MLLM-based Framework for Multimodal Emotion Recognition in Conversation

Multimodal emotion recognition in conversation (MERC), the task of identifying the emotion label for each utterance in a conversation, is vital for developing empathetic machines. Current MLLM-based MERC studies focus mainly on capturing the speaker's textual or vocal characteristics, but ignore the significance of video-derived behavior information. Different from text and audio inputs, learning videos with rich facial expression, body language and posture, provides emotion trigger signals to the models for more accurate emotion predictions. In this paper, we propose a novel behavior-aware MLLM-based framework (BeMERC) to incorporate speaker's behaviors, including subtle facial micro-expression, body language and posture, into a vanilla MLLM-based MERC model, thereby facilitating the modeling of emotional dynamics during a conversation. Furthermore, BeMERC adopts a two-stage instruction tuning strategy to extend the model to the conversations scenario for end-to-end training of a MERC predictor. Experiments demonstrate that BeMERC achieves superior performance than the state-of-the-art methods on two benchmark datasets, and also provides a detailed discussion on the significance of video-derived behavior information in MERC.

CKERC : Joint Large Language Models with Commonsense Knowledge for Emotion Recognition in Conversation

Emotion recognition in conversation (ERC) is a task which predicts the emotion of an utterance in the context of a conversation. It tightly depends on dialogue context, speaker identity information, multiparty dialogue scenario and so on. However, the state-of-the-art method (instructERC) solely identifying speaker, and ignores commonsense knowledge(i.e., reaction of the listeners and intention of the speaker, etc.) behind speakers during a conversation, which can deeply mine speaker information. To this end, we propose a novel joint large language models with commonsense knowledge framework for emotion recognition in conversation, namely CKERC.We design prompts to generate interlocutors' commonsense based on historical utterances with large language model. And we use the interlocutor commonsense identification task for LLM pre-training to fine-tune speaker implicit clues information.By solving above challenge, our method achieve state-of-the-art.We extensive experiment on three widely-used datasets, i.e., IEMOCAP, MELD, EmoryNLP, demonstrate our method superiority. Also, we conduct in-depth analysis and further demonstrate the effectiveness of commonsense knowledge in ERC task in large language model.