United States

Time plays a critical role in how information is generated, retrieved, and interpreted. In this survey, we provide a comprehensive overview of Temporal Question Answering (TQA), a research area that focuses on answering questions involving temporal constraints or context. As time-stamped content from sources like news articles, web archives, and knowledge bases continues to grow, TQA systems must address challenges such as detecting temporal intent, normalizing time expressions, ordering events, and reasoning over evolving or ambiguous facts. We organize existing work through a unified perspective that captures the interaction between corpus temporality, question temporality, and model capabilities, enabling a systematic comparison of datasets, tasks, and approaches. We review recent advances in TQA enabled by neural architectures, especially transformer-based models and Large Language Models (LLMs), highlighting progress in temporal language modeling, retrieval-augmented generation (RAG), and temporal reasoning. We also discuss benchmark datasets and evaluation strategies designed to test temporal robustness, recency awareness, and generalization.

ACL 2026

It&#39;s High Time: A Survey of Temporal Question Answering

### Welcome!
The 62nd Annual Meeting of the Association for Computational Linguistics (ACL 2024) will take place in Bangkok, Thailand from August 11th to 16th, 2024. Our Virtual Poster Sessions will take place online Thursday, August 22, 2024.

You are required to register for this event. **Please register [here](https://2024.aclweb.org/registration). **

If you have already registered, please check your inbox for an email from Underline granting you access to ACL 2024 content.

Please register!

ACL 2024

The 62nd Annual Meeting of the Association for Computational Linguistics (ACL 2024) will take place in Bangkok, Thailand from August 11th to 16th, 2024. More information will be announced soon.

It's High Time: A Survey of Temporal Question Answering

technical paper

It is my great pleasure and honor to welcome you to the 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026), held in San Diego, California, USA, July 2–7, 2026. ACL 2026 continues our fields tradition of excellence in scholarship, innovation, and inclusivity, and I am deeply grateful to the many volunteers who have worked tirelessly to bring this event to life.

It is exciting to see the many developments in the field reflected at our conference, such as the advancement of large language models (born out of our community) into many application directions. This is not without challenges, partly due to unsolved problems, and partly due to misuse of the technology. All this is reflected well at ACL 2026. The conference also reflects that the field is not only about advancing technology, but also the science of fundamentally understanding language and how it is use by many different speaker communities. 
[Full Message from General Chair Philipp Koehn](https://drive.google.com/file/d/1QiJFOmMjvDgsDo3i1XzlQnImxe3RU_Cy/view?usp=sharing) 
<html><button style="display: inline-flex; align-items: center; justify-content: center; white-space: nowrap; border-radius: 9999px; font-weight: bold; background: #7c3aed; color: white; font-family: 'Space Grotesk', sans-serif; height: 40px; font-size: 16px; padding: 0 20px; border: none; cursor: pointer" onclick="window.open('https://drive.google.com/file/d/1QiJFOmMjvDgsDo3i1XzlQnImxe3RU_Cy/view?usp=sharing')">Conference Handbook</button></html>

To access this event page, you need to log in with the **email address you registered with**. Access credentials will be sent to your email from Underline - subject line "Welcome to ACL 2026". Please be sure to check your spam email folder if you do not see an email confirmation right away.

To access this event page, you are required to register.
Please complete your registration to continue.

We recommend reading [**the registration information**](https://2026.aclweb.org/registration/) first.

Registration Required

The 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026) will take place in San Diego, California, from July 2 - 7, 2026.

Evaluating the safety robustness of LLMs is critical for their deployment. However, mainstream Red Teaming methods rely on online generation and black-box output analysis. These approaches are not only costly but also suffer from feedback latency, making them unsuitable for agile diagnostics after training a new model.
To address this, we propose N-GLARE (A Non-Generative, Latent Representation-Efficient LLM Safety Evaluator). N-GLARE operates entirely on the model's latent representations, bypassing the need for full text generation. It characterizes hidden layer dynamics by analyzing the APT (Angular-Probabilistic Trajectory) of latent representations and introducing the JSS (Jensen-Shannon Separability) metric.
Experiments on over 40 models and 20 red teaming strategies demonstrate that the JSS metric exhibits high consistency with Red Teaming safety rankings at less than 1\% token and runtime cost.

N-GLARE: An Non-Generative Latent Representation-Efficient LLM Safety Evaluator

While Large Language Models (LLMs) are frequently cited for their sophisticated pragmatic reasoning \cite{10.5555/3600270.3602070, Bubeck2023SparksOA}, recent progress in sarcasm detection increasingly relies on synthetic benchmarks \cite{li25ba_interspeech, anonymous2025deceptive}. This study exposes a catastrophic generalization gap in this paradigm: we observe that models achieve near-perfect accuracy on synthetic data but collapse to random guessing on organic human speech. By triangulating hidden state geometry, entropy analysis, and causal interventions, we demonstrate that this disparity stems from shortcut learning \cite{Geirhos_2020}—models exploit the low-entropy statistical signatures of generated text while remaining ``semantically blind'' to the pragmatic cues essential for irony. Our findings indicate that high performance on synthetic leaderboards reflects forensic pattern matching rather than the genuine linguistic intelligence assumed in prior work, creating a statistical mirage of competence.

LLMs in Sarcasm Detection? It's elementary! (Or is it?)

As large language models (LLMs) are more frequently used in retrieval-augmented generation pipelines, it is increasingly relevant to study their behavior under knowledge conflicts. Thus far, the role of the source of the retrieved information has gone unexamined. We address this gap with a novel framework to investigate how source preferences affect LLM resolution of inter-context knowledge conflicts in English, motivated by interdisciplinary research on credibility. By using synthetic sources, we study preferences for different types of sources without inheriting the biases of specific real-world sources. With a comprehensive, tightly-controlled evaluation of $13$ open-weight LLMs, we find that LLMs prefer institutionally-corroborated information (e.g., government or newspaper sources) over information from people and social media. However, these source preferences can be reversed by simply repeating information from less credible sources. To mitigate repetition effects and maintain consistent preferences, we propose a novel method that reduces repetition bias by up to $79.2$\%, while also maintaining at least $72.5$\% of original preferences. We release all data and code to encourage future work on credibility and source preferences in knowledge-intensive NLP.

Whose Facts Win? LLM Source Preferences under Knowledge Conflicts

Natural Language to Excel Formula (NL2Formula) translates user intent into executable spreadsheet formulas. However, current models often produce near-miss outputs—formulas that parse correctly yet fail at execution due to an incorrect function, operator, or reference. Through a systematic error analysis, we find that these errors repeatedly arise from a small set of structural decision points, motivating the need for typed error supervision rather than general error signals. To this end, we introduce an abstract syntax tree (AST)-based error taxonomy that organizes common error modes by the kind of decision that goes wrong in the parse tree. Building on this taxonomy, we propose Error-Aware Contrastive Few-Shot Learning (ECFL), an error-aware framework that unifies training and inference around typed error supervision. During offline training, ECFL mines near-miss errors, assigns error types under the taxonomy, and builds error-aware contrastive demonstrations for fine-tuning. During online inference, a lightweight predictor estimates likely error types and triggers targeted retrieval of contrastive demonstrations to guide single-pass decoding. Experiments show ECFL improves Exact Match (EM) by 6.4 points over supervised fine-tuning (SFT) and matches self-consistency (SC@5) accuracy at substantially lower inference cost.

Learning from Near-Misses: Error-Aware Contrastive Few-Shot Learning for NL2Formula

Supervised Fine-Tuning (SFT) is the predominant paradigm for aligning large language models (LLMs), yet it suffers from optimization instability and limited generalization. Recent work attributes this issue to pathological gradient scaling and proposes Dynamic Fine-Tuning (DFT) to correct it at the token level. However, DFT assumes all demonstrations are equally suitable learning targets, an assumption violated by the strong heterogeneity of large-scale instruction data, where demonstration-policy mismatch induces high-variance updates at the sample level. We introduce Compatibility-Aware Dynamic Fine-Tuning (CADFT), a principled extension of DFT that controls sample-level optimization variance. CADFT derives a dynamic, policy-dependent compatibility signal from model likelihoods to modulate supervised updates, suppressing high-variance gradients from incompatible demonstrations. We further propose a delayed, low-frequency compatibility-guided rewriting strategy to transform persistently incompatible demonstrations into learnable targets. We show that CADFT can be interpreted as a variance-controlled estimator that generalizes token-level stabilization in DFT to the sample level. Extensive experiments demonstrate improved stability, generalization, and cold-start reinforcement learning initialization, while remaining fully supervised and free of reward modeling.

Compatibility-Aware Dynamic Fine-Tuning for Large Language Models

Personality disorders (PDs) are a complex class of mental health (MH) conditions characterized by persistent patterns of cognition, behavior, and emotional regulation that deviate from cultural norms. While social media has become a valuable resource for MH research, NLP has largely focused on more prevalent conditions (e.g., depression), leaving PDs underexplored. In this work, we introduce PersonalityDBench, a large-scale, clinically grounded dataset that supports multidimensional study of personality pathology, and standardized, reproducible evaluation of LLM steering toward clinically grounded behavioral targets. The dataset comprises two parts: (1) PRISMA and (2) PersonaDSteering. (1) PRISMA (PeRsonality dISorder MAnifestations) is a clinically annotated collection of social media content spanning the full spectrum of PDs. It links clinically validated diagnostic criteria and dimensional trait frameworks with computational annotation and analysis methods to support fine-grained, multidimensional study of how PDs manifests in naturalistic, free-form language. Building on PRISMA, (2) PersonaDSteering is a benchmark for LLM steering evaluation that operationalizes clinically grounded PD profiles into structured behavioral elicitation tasks, enabling multidimensional steerability assessment beyond single-behavior settings and supporting PD-consistent persona construction for simulated patient generation. This dataset may have application in the study and modeling of PD and powering personality-specific text generation for adaptive, personalized chat systems.

PersonalityDBench: A Dataset for Personality Disorders - from Modeling to Controlled Generation

Text-attributed graphs (TAGs) require jointly modeling relational structure and node-level text. Existing GNN-LLM approaches perform by incorporating large language models at inference time for processing the text attributes, resulting in costly deployment. More fundamentally, LLM knowledge is typically used in a sample-wise manner, leading to inefficient utilization across graph instances. In this work, we study how interactions with LLM embedding spaces affect graph representations, and show that projecting into the LLM space can learn better GNNs. That is to say, the knowledge encoded in LLM embeddings can be compressed into graph representations. Based on this insight, we propose a framework that internalizes LLM knowledge within graph models and supports inference-efficient TAG learning. Our framework employs a hierarchical Proxy-Purifier module with distribution-level regularization, using LLM embeddings only as training-time guidance. With this module, the model operates TAGs without invoking LLMs, achieving high efficiency as standard GNNs without LLMs. Notably, experiments on five popular TAG tasks further demonstrate that our method can also achieve consistent performance gains, in comparison to existing GNN-LLM approaches.

Compressing LLM Knowledge into Graph Representations for Text-attributed Graphs Learning

Emergent misalignment, where fine-tuning on narrow, non-harmful tasks induces harmful behaviors, poses a key challenge for AI safety in LLMs. Despite growing empirical evidence, its underlying mechanism remains unclear. To uncover the reason behind this phenomenon, we propose a mechanistic account based on the geometry of feature superposition. Because features are encoded in overlapping, fine-tuning that amplifies a target feature also unintentionally strengthens nearby harmful features in accordance with their similarity. We give a simple gradient-level derivation of this mechanism and empirically test it across multiple LLMs (Gemma-2 2B/9B/27B, LLaMA-3.1 8B, gpt-oss 20B). Using sparse autoencoders (SAEs), we identify features tied to misalignment-inducing data and to harmful behaviors, and show that they are geometrically closer to each other than features derived from non-inducing data. This trend generalizes across domains (e.g., health, career, legal advice). Finally, we show that a geometry-aware approach—filtering training samples nearest to toxic features—reduces misalignment by 34.5%, substantially outperforming random removal and achieving stronger mitigation than LLM-as-a-judge–based filtering. Our study explains emergent misalignment through feature superposition, providing a basis for understanding and mitigating this phenomenon.

Understanding Emergent Misalignment via Feature Superposition Geometry

The proliferation of Large Language Models (LLMs) has saturated social media platforms with hyper-realistic posts, rendering traditional detection methods that rely on low-level artifacts or unimodal statistics increasingly ineffective. In this work, we identify a fundamental semantic distinction: humans tend to complement visual content with additional context, while LLMs predominantly describe the visual information. To capture this, UMPIRE employs an orthogonal semantic decomposition mechanism that disentangles textual embeddings into redundant and complementary components. An adaptive gating module dynamically weighs these components to reflect diverse communicative styles. To enforce the desired geometric structure, we introduce a latent contrastive redundancy regularization loss that encourages LLM-generated content to exhibit high semantic redundancy, while human-written content emphasizes complementarity. Experimental results demonstrate that UMPIRE significantly outperforms state-of-the-art detection methods across multiple datasets, achieving up to a 5.38% improvement in accuracy.

UMPIRE: Unveiling LLM-generated Posts via Redundant Expressions

Textual humor is enormously diverse and computational studies need to account for this range, including intentionally bad humor. In this paper, we curate and analyze a novel corpus of sentences from the Bulwer-Lytton Fiction Contest to better understand "bad" humor in English. Standard humor detection models perform poorly on our corpus, and an analysis of literary devices finds that these sentences combine features common in existing humor datasets (e.g., puns, irony) with metaphor, metafiction and simile. LLMs prompted to synthesize contest-style sentences imitate the form but exaggerate the effect by over-using certain literary devices, and including far more novel adjective-noun bigrams than human writers.

Downloads

Next from ACL 2026

N-GLARE: An Non-Generative Latent Representation-Efficient LLM Safety Evaluator

Similar lecture

Towards Robust Temporal Reasoning of Large Language Models via a Multi-Hop QA Dataset and Pseudo-Instruction Tuning

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from ACL 2026

N-GLARE: An Non-Generative Latent Representation-Efficient LLM Safety Evaluator

Similar lecture

Towards Robust Temporal Reasoning of Large Language Models via a Multi-Hop QA Dataset and Pseudo-Instruction Tuning

Downloads