Papers by Paolo Papotti
Definitions Matter: Guiding GPT for Multi-label Classification (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Recent success of Large Language Models (LLMs) is due to their superior performance on various tasks such as text generation, summarization, question answering, and inductive reasoning. |
| Approach: | They propose to generate definitions from examples and use them for zero-shot classification and to investigate how an LLM makes use of the definitions. |
| Outcome: | The proposed method improves the definitions of class labels and improves their understanding of the definition. |
Transformers for Tabular Data Representation: A Survey of Models and Applications (2023.tacl-1)
Copied to clipboard
| Challenge: | Recent research efforts extend LMs by developing neural representations for structured data. |
| Approach: | They propose to extend transformer-based language models to tabular data by analyzing inputs, model training, and supported downstream tasks. |
| Outcome: | The proposed models are compared against existing models and are based on a traditional pipeline. |
CacheNotes: Task-Aware Key-Value Cache Compression for Reasoning-Intensive Knowledge Tasks (2026.eacl-long)
Copied to clipboard
| Challenge: | Current methods for integrating external knowledge into Large Language Models (LLMs) face limitations with broad, multi-source queries, while long-context models are computationally prohibitive. |
| Approach: | They propose a task-aware key-value cache compression method that generates a sequence of CPTs from a corpus and guides a one-time compression of the corpus into a compact, reusable KV cache. |
| Outcome: | The proposed method outperforms Retrieval-Augmented Generation (RAG) on Question-Answering tasks and reduces latency by over 4. |
SQUAB: Evaluating LLM robustness to Ambiguous and Unanswerable Questions in Semantic Parsing (2025.emnlp-main)
Copied to clipboard
| Challenge: | Practical user questions often deviate from ideal conditions, challenging the applicability of existing benchmarks. |
| Approach: | They propose an automatic dataset generator of Ambiguous and Unanswerable questions that generates complex, annotated SP tests using a blend of SQL and LLM capabilities. |
| Outcome: | The proposed framework reduces test generation costs by up to 99% while aligning with real-world question patterns. |
Automated Detection of Tropes In Short Texts (2025.coling-main)
Copied to clipboard
| Challenge: | Tropes are often used in movies to convey familiar patterns, but they also play a significant role in online communication . |
| Approach: | They propose to automatically detect tropes in social media posts by using a dataset . they define the task, distinguish it from previous work, and develop a machine learning technique . |
| Outcome: | The proposed method can detect tropes in social media posts with high accuracy. |
RuleBERT: Teaching Soft Rules to Pre-Trained Language Models (2021.emnlp-main)
Copied to clipboard
| Challenge: | Pre-trained language models (PLMs) are limited in their ability to capture and use common-sense knowledge. |
| Approach: | They propose to teach PLMs how to reason with soft Horn rules by leveraging logical rules to learn how to predict precise probabilities. |
| Outcome: | The proposed model performs well on logical rules that were unseen at training. |
You Are My Type! Type Embeddings for Pre-trained Language Models (2022.findings-emnlp)
Copied to clipboard
| Challenge: | Existing work has shown that Pre-trained language models can encode semantic types, but it is not clear how to use types to steer the output. |
| Approach: | They propose to embed a type by a small set of word examples to promote desired types in a PLM. |
| Outcome: | The proposed model can represent types and steer masking predictions without changes to the prompt text without changes in the prompt. |
Refining Attention for Explainable and Noise-Robust Fact-Checking with Transformers (2025.emnlp-main)
Copied to clipboard
| Challenge: | Conventional transformer-based models falter due to noise sensitivity and lack explainability . ATTUN is a transformer architecture designed to enhance model transparency and resilience to noise. |
| Approach: | They propose a transformer architecture that enhances model transparency and resilience to noise . ATTUN is a module that directly modifies attention weights . they validated their approach using fact-checking datasets based on their results . |
| Outcome: | The proposed model improves predictions and identify relevant sections of input data. |
An LLM-Based Approach for Insight Generation in Data Analysis (2025.naacl-long)
Copied to clipboard
| Challenge: | Existing approaches to generate insightful data from databases are time-consuming and resource-intensive. |
| Approach: | They propose a method that leverages Large Language Models to automatically generate textual insights from databases. |
| Outcome: | The proposed approach generates more insightful insights than other approaches while maintaining correctness. |
Unknown Claims: Generation of Fact-Checking Training Examples from Unstructured and Structured Data (2024.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for fact-checking are labor-intensive and time-consuming. |
| Approach: | They propose a framework that generates training instances for FC systems automatically using textual and tabular content. |
| Outcome: | The proposed framework generates training instances for FC systems using textual and tabular content. |
Parallel Context-of-Experts Decoding for Retrieval Augmented Generation (2026.findings-acl)
Copied to clipboard
| Challenge: | Retrieval Augmented Generation relies on concatenating documents into a long context prompt, causing prefill bottlenecks. |
| Approach: | They propose a training-free framework that shifts evidence aggregation from attention to decoding . they treat retrieved documents as isolated "experts", synchronizing their predictions via a retrieval-aware extension of context-awful decoding. |
| Outcome: | The proposed framework shifts evidence aggregation from attention to decoding . it treats retrieved documents as isolated experts, synchronizing their predictions . |