Pasquale Minervini
Lecturer in Natural Language Processing
Researches large language models, including retrieval-augmented generation, model steering, evaluation and knowledge graph reasoning.
Pasquale Minervini's lab at Edinburgh works on knowledge, QA & retrieval, large language models and graph mining & networks. 69 papers since 2023, 36 at top venues, mostly EMNLP and ACL.
At a glance
- papers, 2023–2026
- 18/ year
- at T1 venues
- 52%
- PhD students
- —
- MS students
- —
- alumni listed
- —
Papers by yearfrom the lab homepage, 2023–2026
- 16 papers2026 · T1 9 · collecting
- 26 papers2025 · T1 15
- 18 papers2024 · T1 8
- 9 papers2023 · T1 4
T1 venuesOtherVenues: EMNLP, ACL, NeurIPS, ICML, AAAI
Research focuspaper titles that use each topic's words
- Large language models31 papers
- Knowledge, QA & retrieval13 papers
- Graph mining & networks6 papers
- knowledge graph2 papers
- researches large0 papers
Paper list
From the lab homepage.
202616 papers
- Post-AGI Science and Society WorkshopAgentic Uncertainty Reveals Agentic Overconfidence (opens in a new tab)
- Findings of EACLAnalyzing LLM Instruction Optimization for Tabular Fact Verification (opens in a new tab)
- EMNLP (T1)Can Retrieval Heads See Images? Multimodal Retrieval Heads in Long-Context Vision-Language Models (opens in a new tab)
- NeurIPS (T1)Do Composed Image Retrieval Benchmarks Require Multimodal Composition? (opens in a new tab)
- ICML (T1)Fast and Expressive Multi-Token Prediction with Probabilistic Circuits (opens in a new tab)
- Findings of EMNLPInterference-Aware Composition of Steering Vectors (opens in a new tab)
- ESANNLearning and Reasoning on Knowledge and Heterogeneous Graphs in the era of Graph Foundation and Large Language Models (opens in a new tab)
- ICML (T1)Learning GUI Grounding with Spatial Reasoning from Visual Feedback (opens in a new tab)
- NeurIPS WorkshopLogit-Contribution Scoring Identifies Non-Literal Retrieval Heads (opens in a new tab)
- LRECLow-Rank Compression of Language Models via Differentiable Rank Selection (opens in a new tab)
- ACL (T1)PiCSAR: Probabilistic Confidence Selection And Ranking for Reasoning Chains (opens in a new tab)
- EMNLP (T1)Same Answer, Different Representations: Hidden Instability in VLMs (opens in a new tab)
202526 papers
- AAAI (T1)Adaptive Computation Modules: Granular Conditional Computation for Efficient Inference (opens in a new tab)
- NeurocomputingAdaptive Layer and Token Selection for Efficient Fine-Tuning of Vision Transformers (opens in a new tab)
- ICLR (T1)An Auditing Test to Detect Behavioral Shift in Language Models (opens in a new tab)
- NAACL (T1)Are We Done with MMLU? (opens in a new tab)
- EMNLP (T1)DeCoRe: Decoding by Contrasting Retrieval Heads to Mitigate Hallucinations (opens in a new tab)
- IJCNLP-AACLEnhancing Long Document Long Form Summarisation with Self-Planning (opens in a new tab)
- ESANNEnhancing neural link predictors for temporal knowledge graphs with temporal regularisers (opens in a new tab)
- PKDD/ECML WorkshopsFine-Tuning Foundation Models for Temporal Knowledge Graph Reasoning (opens in a new tab)
- EMNLP (T1)FLARE: Faithful Logic-Aided Reasoning and Exploration (opens in a new tab)
- EMNLP (T1)GRADA: Graph-based Reranking against Adversarial Documents Attack (opens in a new tab)
- TMLR (T1)Inverse Scaling in Test-Time Compute (opens in a new tab)
- International Conference on Machine Learning (ICML ) Spotlight, top 2.6% of accepted papersIs Complex Query Answering Really Complex? (opens in a new tab)
202418 papers
- EMNLP (T1)A Simple and Effective $L_{2}$ Norm-Based Strategy for KV Cache Compression (opens in a new tab)
- ACL (T1)Analysing The Impact of Sequence Composition on Language Model Pre-Training (opens in a new tab)
- EMNLP (T1)Atomic Inference for NLI with Generated Facts as Atoms (opens in a new tab)
- ICML WorkshopAttention Is All You Need But You Don’t Need All Of It For Inference of Large Language Models (opens in a new tab)
- ClinicalNLPCan LLMs Correct Physicians, Yet? Investigating Effective Interaction Methods in the Medical Domain (opens in a new tab)
- Intelligenza Artificiale 18(1): 175-190Conditional computation in neural networks: principles and research trends (opens in a new tab)
- ClinicalNLPEdinburgh Clinical NLP at MEDIQA-CORR 2024: Guiding Large Language Models with Hints (opens in a new tab)
- ClinicalNLPEdinburgh Clinical NLP at SemEval-2024 Task 2: Fine-tune your model unless you have access to GPT-4 (opens in a new tab)
- Conference on Language ModelingEvaluating the Adversarial Robustness of Retrieval-Based In-Context Learning for Large Language Models (opens in a new tab)
- Conference on Language ModelingForklift: An Extensible Neural Lifter (opens in a new tab)
- Frontiers in Artificial Intelligence 7, 1464690Fostering Effective Hybrid Human-LLM Reasoning and Decision Making (opens in a new tab)
- Nature Human BehaviourLarge language models surpass human experts in predicting neuroscience results (opens in a new tab)
20239 papers
- NeurIPS (T1)Adapting Neural Link Predictors for Data-Efficient Complex Query Answering (opens in a new tab)
- AAAI (T1)Adaptive Perturbation-Based Gradient Estimation for Discrete Latent Variable Models (opens in a new tab)
- Chapter in Compendium of Neurosymbolic Artificial Intelligence, IOS PressApproximate Answering of Graph Queries (opens in a new tab)
- CIKM (T1)Combining Inductive and Deductive Reasoning for Query Answering over Incomplete Knowledge Graphs (opens in a new tab)
- IJCNNLearning Survival Models for Relapse Prediction in Early-Stage Lung Cancer Patients (opens in a new tab)
- JCO Clinical Cancer Informatics :7Machine Learning–Assisted Recurrence Prediction for Patients With Early-Stage Non–Small-Cell Lung Cancer (opens in a new tab)
- NeurIPS (T1)No Train No Gain: Revisiting Efficient Training Algorithms For Transformer-based Language Models (opens in a new tab)
- CoNLLREFER: An End-to-end Rationale Extraction Framework for Explanation Regularization (opens in a new tab)
- EACLXQA-DST: Multi-Domain and Multi-Lingual Dialogue State Tracking (opens in a new tab)
Where alumni went
Alumni are not listed on the lab homepage.
Apply to EdinburghFall 2027
Not announced yet
Informatics PhD
- Round
- Rolling (funding consideration)
- Opens
- —
- GRE
- See program page
- English
- See program page
- Fee
- none
Not yet confirmed for this intake.
Similar labsnearest by research summary
- GBEdinburgh
Ivan Titov
- Large language models
- USUCLA
Vwani Roychowdhury
- Language & society
- HKCUHK
Wai Lam
- Knowledge, QA & retrieval
- CNTsinghua
Qingyao Ai
- Large language models
- GBEdinburgh
Wenda Li
- Reasoning & neuro-symbolic AI
- USNYU
Ernest Davis
- Reasoning & neuro-symbolic AI
Your notes
Save this lab to keep notes.
Data
Some numbers could not be verified:
- The member list could not be found on the lab site.
- The alumni list could not be found on the lab site.
- Collected
- Verified
- Papers
- The lab homepage
- Members
- Not collected
- Alumni
- Not collected
Tiers are venue classifications, not lab rankings.