Related papers: C$^{3}$T: Counterfactual Causal Reasoning for Sentiment Shifts in Social-Media Conversation Trees

C$^{3}$T: Counterfactual Causal Reasoning for Sentiment Shifts in Social-Media Conversation Trees

URL: http://arxiv.org/abs/2609.02131v1
Date: Wed, 02 Sep 2026 05:40:06 GMT
Title: C$^{3}$T: Counterfactual Causal Reasoning for Sentiment Shifts in Social-Media Conversation Trees
Abstract summary: We study why sentiment changes in rumor-centric conversation trees by treating discourse moves as candidate interventions.<n>We propose C$3$T (Counterfactual Causal Conversation Transformer), a thread-structured temporal model that jointly predicts node sentiment and shifts.
Score: 0.3093890460224435
License: http://creativecommons.org/licenses/by/4.0/
Abstract: Sentiment in social-media threads does not only vary across posts; it shifts as users react to claims, corrections, evidence, and hostility within a branching reply tree. We study why sentiment changes in rumor-centric conversation trees by treating discourse moves (e.g., denial/correction, evidence/link, toxicity/attack) as candidate interventions and asking (i) what sentiment a reply expresses, (ii) whether the sentiment shifts relative to its parent, and (iii) which prior message most plausibly drove the reply's sentiment. To support this setting, we introduce CaSiRe, a causal sentiment reasoning layer over public rumor conversation datasets that adds post-level sentiment labels, induced parent-child shift labels, calibrated multi-label intervention tags, and explicitly annotated causal-source labels. We then propose C$^{3}$T (Counterfactual Causal Conversation Transformer), a thread-structured temporal model that jointly predicts node sentiment and shifts, learns sparse ancestor attribution, and supports counterfactual queries by forcing conversational intervention embeddings on or off to estimate potential outcomes. Under an event-level split, C$^{3}$T improves out-of-event robustness and attribution over text-only, graph-based, and temporal baselines, and yields interpretable model-based effects: denials/corrections and evidence reduce downstream negativity, while toxicity increases it. We also benchmark open-weight LLM prompting baselines and find that added conversational context helps, but attribution remains less reliable, motivating structure-aware counterfactual modeling for social-media analysis.

Related papers

A Dynamic-Semantics Framework for Grounding Human Referring Expressions in Visual Perceptual Data [0.0]
We present a framework that externalizes pact state into three explicit, inspectable sets of referent-object bindings.<n>We evaluate the framework on the Stanford Repeated Reference Game corpus.
arXiv Detail & Related papers (2026-08-09T12:14:10Z)
When to Think, When to Speak: Learning Disclosure Policies for LLM Reasoning [60.714416943748866]
In single-stream autoregressive interfaces, the same tokens both update the model state and constitute an irreversible public commitment.<n>We introduce Side-by-Side (SxS) Interleaved Reasoning, which makes disclosure timing a controllable decision within standard autoregressive generation.
arXiv Detail & Related papers (2026-05-05T02:59:58Z)
ControBench: An Interaction-Aware Benchmark for Controversial Discourse Analysis on Social Networks [17.87131765035957]
ControBench is a benchmark for controversial discourse analysis.<n>It is built from Reddit discussions on Trump, abortion, and religion.<n>We evaluate graph neural networks, pretrained language models, and large language models on ControBench.
arXiv Detail & Related papers (2026-05-01T08:39:26Z)
Real Talk, Virtual Faces: A Formal Concept Analysis of Personality and Sentiment in Influencer Audiences [0.981328290471248]
We propose a two-layer framework grounded in Formal Concept Analysis and association rule mining.<n>Applying to YouTube comments from three VI--HI influencer pairs, the two-layer analysis reveals a consistent structural divergence.<n>Our results position FCA as a principled tool for multi-signal discourse analysis.
arXiv Detail & Related papers (2026-03-25T15:23:32Z)
CausalFlip: A Benchmark for LLM Causal Judgment Beyond Semantic Matching [50.65932158912512]
We propose a new causal reasoning benchmark, CausalFlip, to encourage the development of new large language models.<n>CaulFlip consists of causal judgment questions built over event triples that could form different confounder, chain, and collider relations.<n>We evaluate LLMs under multiple training paradigms, including answer-only training, explicit Chain-of-Thought supervision, and a proposed internalized causal reasoning approach.
arXiv Detail & Related papers (2026-02-23T18:06:15Z)
ADEPT: RL-Aligned Agentic Decoding of Emotion via Evidence Probing Tools -- From Consensus Learning to Ambiguity-Driven Emotion Reasoning [67.22219034602514]
We introduce ADEPT (Agentic Decoding of Emotion via Evidence Probing Tools), a framework that reframes emotion recognition as a multi-turn inquiry process.<n> ADEPT transforms an SLLM into an agent that maintains an evolving candidate emotion set and adaptively invokes dedicated semantic and acoustic probing tools.<n>We show that ADEPT improves primary emotion accuracy in most settings while substantially improving minor emotion characterization.
arXiv Detail & Related papers (2026-02-13T08:33:37Z)
Importance-aware Topic Modeling for Discovering Public Transit Risk from Noisy Social Media [8.638879065913246]
Urban transit agencies increasingly turn to social media to monitor emerging service risks such as crowding, delays, and safety incidents.<n>We address this challenge by jointly modeling linguistic interactions and user influence.<n>The proposed model achieves state-of-the-art topic coherence and strong diversity compared with leading baselines.
arXiv Detail & Related papers (2025-12-06T04:45:17Z)
Boosting Temporal Sentence Grounding via Causal Inference [55.61521060331558]
Temporal Sentence Grounding aims to identify relevant moments in an untrimmed video that semantically correspond to a given textual query.<n>These spurious correlations arise from two primary factors: (1) inherent biases in the textual data, such as frequent co-occurrences of specific verbs or phrases, and (2) the model's tendency to overfit to salient or repetitive patterns in video content.<n>We propose a novel TSG framework, causal intervention and counterfactual reasoning that utilize causal inference to eliminate spurious correlations and enhance the model's robustness.
arXiv Detail & Related papers (2025-07-07T13:01:06Z)
What Causes the Failure of Explicit to Implicit Discourse Relation Recognition? [14.021169977926265]
We show that one cause for such failure is a label shift after connectives are eliminated. We find that the discourse relations expressed by some explicit instances will change when connectives disappear. We investigate two strategies to mitigate the label shift: filtering out noisy data and joint learning with connectives.
arXiv Detail & Related papers (2024-04-01T09:08:53Z)
Neutral Utterances are Also Causes: Enhancing Conversational Causal Emotion Entailment with Social Commonsense Knowledge [52.04642421708207]
Conversational Causal Emotion Entailment aims to detect causal utterances for a non-neutral targeted utterance from a conversation. Emotion information can markedly promote the detection of causal utterances whose emotion is the same as the targeted utterance.
arXiv Detail & Related papers (2022-05-02T09:12:32Z)
Information Consumption and Social Response in a Segregated Environment: the Case of Gab [74.5095691235917]
This work provides a characterization of the interaction patterns within Gab around the COVID-19 topic. We find that there are no strong statistical differences in the social response to questionable and reliable content. Our results provide insights toward the understanding of coordinated inauthentic behavior and on the early-warning of information operation.
arXiv Detail & Related papers (2020-06-03T11:34:25Z)

This list is automatically generated from the titles and abstracts of the papers in this site.