Search Results for author:
Found 13 papers, 5 papers with code
Generative Information Retrieval Evaluation
We resolve this apparent circularity in two ways: 1) by viewing LLM-based assessment as a form of "slow search", where a slower IR system is used for evaluation and training of a faster production IR system; and 2) by recognizing a continuing need to ground evaluation in human assessment, even if the characteristics of that human assessment must change.
A Comparison of Methods for Evaluating Generative IR
Given that Gen-IR systems do not generate responses from a fixed set, we assume that methods for Gen-IR evaluation must largely depend on LLM-generated labels.
Fréchet Distance for Offline Evaluation of Information Retrieval Systems with Sparse Labels
The rapid advancement of natural language processing, information retrieval (IR), computer vision, and other technologies has presented significant challenges in evaluating the performance of these systems.
Adapting Standard Retrieval Benchmarks to Evaluate Generated Answers
In the second, we compare generated answers to the top results retrieved by a diverse set of retrieval models, ranging from traditional approaches to advanced methods, allowing us to measure improvements without human judgments.
Retrieving Supporting Evidence for Generative Question Answering
After presenting a question to an LLM and receiving a generated answer, we query the corpus with the combination of the question + generated answer.
Generative Question Answering
Open-Domain Question Answering
+1
Retrieving Supporting Evidence for LLMs Generated Answers
After presenting a question to an LLM and receiving a generated answer, we query the corpus with the combination of the question + generated answer.
Evaluating Open-Domain Question Answering in the Era of Large Language Models
The recent success of large language models (LLMs) for QA aggravates lexical matching failures since candidate answers become longer, thereby making matching with the gold answers even more challenging.
Early Stage Sparse Retrieval with Entity Linking
In this work, we propose boosting the performance of sparse retrievers by expanding both the queries and the documents with linked entities in two formats for the entity names: 1) explicit and 2) hashed.
Unsupervised Question Clarity Prediction Through Retrieved Item Coherency
While the research community has paid substantial attention to the problem of predicting query ambiguity in traditional search contexts, researchers have paid relatively little attention to predicting when this ambiguity is sufficient to warrant clarification in the context of conversational systems.
Human Preferences as Dueling Bandits
Based on these simulations, one algorithm stands out for its potential.
Predicting Efficiency/Effectiveness Trade-offs for Dense vs. Sparse Retrieval Strategy Selection
These hybrid retrievers leverage low-cost, exact-matching based sparse retrievers along with dense retrievers to bridge the semantic gaps between query and documents.
Shallow pooling for sparse labels
To test this observation, we employed crowdsourced workers to make preference judgments between the top item returned by a modern neural ranking stack and a judged relevant item.
Assessing top-$k$ preferences
To measure the performance of a ranker, we compare its output to this preferred ordering by applying a rank similarity measure. We demonstrate the practical feasibility of this approach by crowdsourcing partial preferences for the TREC 2019 Conversational Assistance Track, replacing NDCG with a new measure named "compatibility".
