The Stochastic Parrot on LLM's Shoulder: A Summative Assessment of Physical Concept Understanding

In a systematic way, we investigate a widely asked question: Do LLMs really understand what they say?, which relates to the more familiar term Stochastic Parrot. To this end, we propose a summative assessment over a carefully designed physical concept understanding task, PhysiCo. Our task alleviates the memorization issue via the usage of grid-format inputs that abstractly describe physical phenomena. The grids represents varying levels of understanding, from the core phenomenon, application examples to analogies to other abstract patterns in the grid world. A comprehensive study on our task demonstrates: (1) state-of-the-art LLMs, including GPT-4o, o1 and Gemini 2.0 flash thinking, lag behind humans by ~40%; (2) the stochastic parrot phenomenon is present in LLMs, as they fail on our grid task but can describe and recognize the same concepts well in natural language; (3) our task challenges the LLMs due to intrinsic difficulties rather than the unfamiliar grid format, as in-context learning and fine-tuning on same formatted data added little to their performance.

The Essence of Contextual Understanding in Theory of Mind: A Study on Question Answering with Story Characters

Theory-of-Mind (ToM) is a fundamental psychological capability that allows humans to understand and interpret the mental states of others. Humans infer others' thoughts by integrating causal cues and indirect clues from broad contextual information, often derived from past interactions. In other words, human ToM heavily relies on the understanding about the backgrounds and life stories of others. Unfortunately, this aspect is largely overlooked in existing benchmarks for evaluating machines' ToM capabilities, due to their usage of short narratives without global backgrounds. In this paper, we verify the importance of understanding long personal backgrounds in ToM and assess the performance of LLMs in such realistic evaluation scenarios. To achieve this, we introduce a novel benchmark, CharToM-QA, comprising 1,035 ToM questions based on characters from classic novels. Our human study reveals a significant disparity in performance: the same group of educated participants performs dramatically better when they have read the novels compared to when they have not. In parallel, our experiments on state-of-the-art LLMs, including the very recent o1 model, show that LLMs still perform notably worse than humans, despite that they have seen these stories during pre-training. This highlights the limitations of current LLMs in capturing the nuanced contextual information required for ToM reasoning.

NeuralMAG: Fast and Generalizable Micromagnetic Simulation with Deep Neural Nets

Micromagnetics has made significant strides, particularly due to its wide-ranging applications in magnetic storage design. Numerical simulation is a cornerstone of micromagnetics research, relying on first-principle rules to compute the dynamic evolution of micromagnetic systems based on the renowned LLG equation, named after Landau, Lifshitz, and Gilbert. However, simulations are often hindered by their slow speed. Although Fast-Fourier transformation (FFT) calculations reduce the computational complexity to O(NlogN), it remains impractical for large-scale simulations. In this paper, we introduce NeuralMAG, a deep learning approach to micromagnetic simulation. Our approach follows the LLG iterative framework but accelerates demagnetizing field computation through the employment of a U-shaped neural network (Unet). The Unet architecture comprises an encoder that extracts aggregated spins at various scales and learns the local interaction at each scale, followed by a decoder that accumulates the local interactions at different scales to approximate the global convolution. This divide-and-accumulate scheme achieves a time complexity of O(N), significantly enhancing the speed and feasibility of large-scale simulations. Unlike existing neural methods, NeuralMAG concentrates on the core computation rather than an end-to-end approximation for a specific task, making it inherently generalizable. To validate the new approach, we trained a single model and evaluated it on two micromagnetics tasks with various sample sizes, shapes, and material settings.

Think out Loud: Emotion Deducing Explanation in Dialogues

Humans convey emotions through daily dialogues, making emotion understanding a crucial step of affective intelligence. To understand emotions in dialogues, machines are asked to recognize the emotion for an utterance (Emotion Recognition in Dialogues, ERD); based on the emotion, then find causal utterances for the emotion (Emotion Cause Extraction in Dialogues, ECED). The setting of the two tasks requires first ERD and then ECED, ignoring the mutual complement between emotion and cause. To fix this, some new tasks are proposed to extract them simultaneously. Although the current research on these tasks has excellent achievements, simply identifying emotion-related factors by classification modeling lacks realizing the specific thinking process of causes stimulating the emotion in an explainable way. This thinking process especially reflected in the reasoning ability of Large Language Models (LLMs) is under-explored. To this end, we propose a new task "Emotion Deducing Explanation in Dialogues" (EDEN). EDEN recognizes emotion and causes in an explicitly thinking way. That is, models need to generate an explanation text, which first summarizes the causes; analyzes the inner activities of the speakers triggered by the causes using common sense; then guesses the emotion accordingly. To support the study of EDEN, based on the existing resources in ECED, we construct two EDEN datasets by human effort. We further evaluate different models on EDEN and find that LLMs are more competent than conventional PLMs. Besides, EDEN can help LLMs achieve better recognition of emotions and causes, which explores a new research direction of explainable emotion understanding in dialogues.

Graph Representation of Narrative Context: Coherence Dependency via Retrospective Questions

This work introduces a novel and practical paradigm for narrative comprehension, stemming from the observation that individual passages within narratives are often cohesively related than being isolated. We therefore propose to formulate a graph upon narratives dubbed NARCO that depicts a task-agnostic coherence dependency of the entire context. Especially, edges in NARCO encompass retrospective free-form questions between two context snippets reflecting high-level coherent relations, inspired by the cognitive perception of humans who constantly reinstate relevant events from prior context. Importantly, our graph is instantiated through our designed two-stage LLM prompting, thereby without reliance on human annotations. We present three unique studies on its practical utility, examining the edge efficacy via recap identification, local context augmentation via plot retrieval, and broader applications exemplified by long document QA. Experiments suggest that our approaches leveraging NARCO yield performance boost across all three tasks.

Previously on the Stories: Recap Snippet Identification for Story Reading

Similar to the "previously-on" scenes in TV shows, recaps can help book reading by recalling the readers' memory about the important elements in previous texts to better understand the ongoing plot. Despite its usefulness, this application has not been well studied in the NLP community. We propose the first benchmark on this useful task called Recap Snippet Identification with a hand-crafted evaluation dataset. Our experiments show that the proposed task is challenging to PLMs, LLMs, and proposed methods as the task requires a deep understanding of the plot correlation between snippets.

SIG: Speaker Identification in Literature via Prompt-Based Generation

Identifying speakers of quotations in narratives is an important task in literary analysis, with challenging scenarios including the out-of-domain inference for unseen speakers, and non-explicit cases where there are no speaker mentions in surrounding context. In this work, we propose a simple and effective approach SIG, a generation-based method that verbalizes the task and quotation input based on designed prompt templates, which also enables easy integration of other auxiliary tasks that further bolster the speaker identification performance. The prediction can either come from direct generation by the model, or be determined by the highest generation probability of each speaker candidate. Based on our approach design, SIG supports out-of-domain evaluation, and achieves open-world classification paradigm that is able to accept any forms of candidate input. We perform both cross-domain evaluation and in-domain evaluation on PDNC, the largest dataset of this task, where empirical results suggest that SIG outperforms previous baselines of complicated designs, as well as the zero-shot ChatGPT, especially excelling at those hard non-explicit scenarios by up to 17% improvement. Additional experiments on another dataset WP further corroborate the efficacy of SIG.

Enhancing Empathetic and Emotion Support Dialogue Generation with Prophetic Commonsense Inference

The interest in Empathetic and Emotional Support conversations among the public has significantly increased. To offer more sensitive and understanding responses, leveraging commonsense knowledge has become a common strategy to better understand psychological aspects and causality. However, such commonsense inferences can be out of context and unable to predict upcoming dialogue themes, resulting in responses that lack coherence and empathy. To remedy this issue, we present Prophetic Commonsense Inference, an innovative paradigm for inferring commonsense knowledge. By harnessing the capabilities of Large Language Models in understanding dialogue and making commonsense deductions, we train tunable models to bridge the gap between past and potential future dialogues. Extensive experiments conducted on EmpatheticDialogues and Emotion Support Conversation show that equipping dialogue agents with our proposed prophetic commonsense inference significantly enhances the quality of their responses.

Plot Retrieval as an Assessment of Abstract Semantic Association

Retrieving relevant plots from the book for a query is a critical task, which can improve the reading experience and efficiency of readers. Readers usually only give an abstract and vague description as the query based on their own understanding, summaries, or speculations of the plot, which requires the retrieval model to have a strong ability to estimate the abstract semantic associations between the query and candidate plots. However, existing information retrieval (IR) datasets cannot reflect this ability well. In this paper, we propose Plot Retrieval, a labeled dataset to train and evaluate the performance of IR models on the novel task Plot Retrieval. Text pairs in Plot Retrieval have less word overlap and more abstract semantic association, which can reflect the ability of the IR models to estimate the abstract semantic association, rather than just traditional lexical or semantic matching. Extensive experiments across various lexical retrieval, sparse retrieval, dense retrieval, and cross-encoder methods compared with human studies on Plot Retrieval show current IR models still struggle in capturing abstract semantic association between texts. Plot Retrieval can be the benchmark for further research on the semantic association modeling ability of IR models.

Multi-level Adaptive Contrastive Learning for Knowledge Internalization in Dialogue Generation

Knowledge-grounded dialogue generation aims to mitigate the issue of text degeneration by incorporating external knowledge to supplement the context. However, the model often fails to internalize this information into responses in a human-like manner. Instead, it simply inserts segments of the provided knowledge into generic responses. As a result, the generated responses tend to be tedious, incoherent, and in lack of interactivity which means the degeneration problem is still unsolved. In this work, we first find that such copying-style degeneration is primarily due to the weak likelihood objective, which allows the model to "cheat" the objective by merely duplicating knowledge segments in a superficial pattern matching based on overlap. To overcome this challenge, we then propose a Multi-level Adaptive Contrastive Learning (MACL) framework that dynamically samples negative examples and subsequently penalizes degeneration behaviors at both the token-level and sequence-level. Extensive experiments on the WoW dataset demonstrate the effectiveness of our approach across various pre-trained models.