Guiding Retrieval using LLM-based Listwise Rankers

Avishek Anand; Mandeep Rathee; Sean MacAvaney

REVIEW 3 major objections 5 minor 39 references

Guiding Retrieval using LLM-based Listwise Rankers

T0 review · 3 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper proposes SlideGar, an adaptive-retrieval loop that lets LLM listwise rerankers pull in documents beyond the initial retrieval pool, improving recall and ranking quality with no increase in the number of LLM inference calls.

desk verdict Sensible extension of graph-based adaptive retrieval to listwise LLM rankers with solid experiments, but the final output ordering is underspecified and needs confirmation from the released code. read the letter →

arxiv 2501.09186 v1 pith:2MTTCNKN submitted 2025-01-15 cs.IR cs.AI

Mandeep Rathee , Sean MacAvaney , Avishek Anand This is my paper · ORCID

classification cs.IRcs.AI

keywords RerankingAdaptiveRetrievalLLMrankersListwiserankingBoundedrecallCorpusgraphPseudo-relevancefeedbackSlidingwindow

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Cascading search pipelines rank whatever the first-stage retriever returns, so relevant documents that miss that cut are gone for good. The paper claims this bounded recall problem can be overcome for LLM listwise rerankers, which output only a relative order and cannot be treated as pointwise scorers. SlideGar adapts graph-based adaptive retrieval to this setting by using each ranked window as feedback: the top documents of a window seed a search of a precomputed corpus graph, and the next window mixes the initial ranking with the graph frontier. On TREC Deep Learning benchmarks, the authors report nDCG@10 gains up to 13.23 percent and recall gains up to 28 percent, at the same LLM inference budget and with about 0.02 percent added latency.

What carries the argument

The central mechanism is the sliding window with graph feedback, given in Algorithm 1. A window of size w is reranked by the listwise LLM; the top b documents are carried into the next window, the remaining documents are appended to the reranked pool, and the graph frontier is replaced with the k-nearest neighbours of the ranked batch, excluding documents already in the pool. The algorithm then alternates the source of the next b documents between the initial ranking and the frontier, so the LLM always sees a mixture of first-stage results and graph-discovered candidates. The corpus graph, built offline from sparse or dense document similarities, supplies the feedback in constant time, and reciprocal rank acts as the pseudo-score that orders the neighbours. This loop keeps the number of LLM calls identical to standard windowed listwise reranking: (c−w)/b + 1 windows for a reranking budget c.

What would settle it

Take a query set and remove from the corpus graph every edge that connects a relevant document to any document in the initial retrieval pool, then run SlideGar: if recall@c still improves over the non-adaptive baseline, the reported gains do not come from the graph edges the paper says they come from. A cheaper check is to replace the graph neighbours with randomly chosen documents and measure whether the recall improvement collapses.

Watch

Extended reading notes

Core claim

SlideGar is a sliding-window variant of graph-based adaptive retrieval designed for listwise LLM rankers. In each step, the LLM reranks a window of w documents; the top b become the seed for the next window, the remainder join the final pool, and the graph frontier is updated with the neighbors of the ranked batch. The algorithm alternates between the initial retrieval pool and the frontier, so the final pool contains documents that never appeared in the first-stage results. Because online listwise rankers give no scores, SlideGar uses reciprocal rank as a pseudo-score to prioritise graph neighbours. The authors demonstrate across four LLM rankers, two first-stage retrievers, and four TREC Deep Learning collections that this loop improves recall and nDCG, and attribute the gains to the Clustering Hypothesis encoded in the corpus graph.

Load-bearing premise

The corpus graph must connect documents the LLM ranks near the top of a window to relevant documents that the initial retrieval missed; if those edges are absent, the feedback loop cannot recover lost relevant documents and the reported recall gains disappear.

Editorial extensions

If this is right

First-stage retrieval no longer sets a hard ceiling on the final ranking: documents the LLM finds relevant can enter the reranked pool via the corpus graph, so weaker lexical first stages can still feed strong LLM rankers.
The recall gain comes at no extra LLM inference cost: the total number of windows, and therefore LLM calls, is unchanged, and the adaptive overhead is a constant-time graph lookup plus a few list operations.
The improvement is consistent across zero-shot and fine-tuned listwise rankers, across sparse and dense first-stage retrievers, and across two different corpora (MSMARCO v1 and v2 with TREC DL19/20/21/22).
A dense-similarity corpus graph consistently outperforms both a sparse-similarity graph and an RM3-based feedback baseline, indicating that the adaptive loop benefits most from complementary dense signals.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

Because SlideGar uses only the LLM's relative order within each window, its seed selection might be improved by exploiting confidence or generation scores from the ranker, a direction the paper does not explore.
The method should degrade gracefully with graph quality, so replacing the static sparse/dense graph with a graph built from the LLM's own document representations could push recall further, at the cost of building that graph.
A stress test for the Clustering Hypothesis: on queries whose relevant documents are topically scattered, such as multi-faceted information needs, the graph-neighbour feedback should contribute little, and the recall gain should shrink accordingly.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

The one thing to know: SlideGar is a genuine and sensible adaptation of Gar/Quam to listwise LLM rerankers, and the experiments are broad and mostly well-executed. The main hang-up is that Algorithm 1 never specifies how the final pool R1 is ordered when computing nDCG@10. If the code sorts by the reciprocal-rank pseudo-scores, fine; if it uses insertion order from the sliding-window alternation, the headline nDCG gains might be an artifact of window scheduling rather than the adaptive feedback itself.

What's new: prior adaptive retrieval required pointwise scores; here they convert listwise output to reciprocal-rank pseudo-scores to prioritize graph-neighbor exploration, and they carry the top-b documents of each window forward as seeds. That combination is, as far as I can tell, unexplored in the cited literature. The empirical coverage is strong: 4 listwise rankers, 2 initial retrievers, 2 corpora, 4 TREC Deep Learning test sets, with significance tests and a k-ablation (Fig. 2) that gives a reasonable check on the clustering hypothesis. The released code, corpus graphs, and de-duplicated indexes are a real plus.

The soft spots, in proportion: the biggest is the unresolved ordering of R1. The stress-test note is right. The paper describes pseudo-scores only for neighbor selection, not for constructing the final ranked list. nDCG@10 is order-dependent, so this is a load-bearing omission for the primary reported metric. Also, the abstract says SlideGar "overcoming the bounded recall problem for LLM rerankers," which is too strong for dense first-stage setups: on TCT-initiated runs the recall gains are smaller and sometimes not significant (e.g., DL20 TCT»RankVicuna). The RM3 baseline is reasonable but not a strong PRF comparator; it mostly serves to show graph-based feedback behaves differently from query expansion.

Who this is for: people working on LLM-based reranking or adaptive retrieval. It deserves a serious referee. I would send it to review, but the authors must clarify the final ranking construction, ideally by pointing to the exact lines in their released code. If the code sorts R1 by reciprocal-rank pseudo-scores, the central claim is likely sound; until then, the paper is incomplete as written.

Referee Report

3 major / 5 minor

Summary. The paper presents SlideGar, an adaptive-retrieval method that extends graph-based adaptive reranking (Gar/Quam) to listwise LLM rerankers. SlideGar processes a sliding window of w documents, carries the top-b documents into the next window, adds the remaining documents to an output pool R1, and alternates the source of the other b documents in the next window between the initial retrieval list and a corpus-graph frontier. The authors evaluate SlideGar with BM25 and TCT first-stage retrieval, four listwise rankers (RankGPT-4o, RankGPT-3.5, RankZephyr, RankVicuna), and the TREC DL19-22 test collections, reporting Recall@c and nDCG@10 gains (up to 28.02% and 13.23% respectively) at a fixed number of LLM calls, together with a graph-depth ablation and a latency measurement.

Significance. If the effectiveness results are reproduced as stated, this is a useful and timely extension of adaptive retrieval to listwise rerankers: it addresses the bounded-recall problem without increasing the number of LLM calls, and the experimental coverage is broad (two corpora, four test sets, three graph substrates, four rankers, two budgets). The paper is also careful in several respects: hyperparameters w=20, b=10, k=16 are taken from prior work rather than tuned on the test sets; significance tests with Bonferroni correction are reported; the graph-depth ablation addresses the dependence on the clustering hypothesis; and the authors release code, indexes, and corpus graphs. The central caveat is that nDCG@10, one of the two headline metrics, depends on the order of the output pool R1, and Algorithm 1 does not specify that order.

major comments (3)

[Section 3.2, Algorithm 1 (lines 9 and 15)] The output R1 is defined through set unions, so its order is unspecified. nDCG@10 is order-dependent, and the claimed nDCG improvements (up to 13.23%, abstract and Table 1) therefore are not well-defined by the paper as written. Please state explicitly whether R1 is an ordered list; if so, specify exactly how B[b+1:|B|] is appended for each window, how the final L1 is placed on top, and whether any final ordering (e.g., by reciprocal-rank pseudo-scores) is applied. If the released code uses a particular ordering, the algorithm description must match it. Without this, the nDCG@10 results cannot be verified.
[Section 3.2, Algorithm 1 line 10 vs. Figure 1 and abstract] Line 10 computes F <- Neighbours(B,G) \ (R1 union L1) over all documents B in the window, whereas the abstract and Figure 1 describe feedback from the most relevant documents seen up to that point, i.e., the top-b carried documents L1. This is a substantive discrepancy: including neighbors of low-ranked documents that were just added to R1 changes which documents can enter the frontier. Please either restrict the frontier update to L1 or redefine Neighbours(B,G) as the neighbors of all documents in the window, and in the latter case specify how neighbors from multiple source documents are prioritized and tie-broken.
[Section 5.1, Table 1] The statement that with the dense first-stage retriever we continue to see improvements is not supported by the table. For example, for TCT»RankGPT-3.5 on DL20, SlideGarTCT lowers nDCG@10 from 0.736 to 0.697 at c=50 and from 0.738 to 0.680 at c=100; similar degradations appear for TCT»RankZephyr and TCT»RankVicuna in several cells. The paper should quantify how often SlideGar helps versus hurts nDCG@10, report significance of degradations, and discuss the conditions under which adaptive expansion is detrimental when the first stage is already strong.

minor comments (5)

[Algorithm 1, line 6] The phrase subject to w and b is vague; please state that B is the permutation of L produced by the listwise ranker under the given window and step parameters.
[Table 1] Significance markers are placed inconsistently (e.g., 0.755 * 0.733 and 0.680 *0.729); please format markers so each metric's significance is unambiguous.
[Section 5.4] Clarify whether the ± values are standard deviations over the five runs; the units and computation should be stated.
[Section 4.1 and Section 2.2] Change de-duplicate to deduplicate, and fix the typo perfomring in Section 2.2; a general language-polish pass is needed.
[Algorithm 1] Specify the behavior when the frontier F contains fewer than b neighbors, since Select(top b from P) would otherwise be undefined.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: SlideGar's claimed gains are evaluated against held-out TREC qrels and depend on external graph/LLM behavior, not on fitted constants or self-referential definitions.

full rationale

The paper's central claim is that SlideGar improves nDCG@10 and recall for listwise LLM rerankers by merging initial-ranking documents with graph-neighborhood documents. The evaluation is measured against TREC DL19-DL22 qrels, which are external to the method, and the reported gains vary by ranker, graph type, and dataset rather than being forced by construction. No parameter is fitted to the test labels: the window size (w=20), step size (b=10), and graph depth (k=16) are fixed defaults inherited from prior work or stated hyperparameters, and the k-ablation in Figure 2 shows sensitivity rather than a tuned-to-test value. The load-bearing assumption, that corpus-graph edges connect LLM-ranked seeds to unretrieved relevant documents, is an empirical premise tested through held-out recall improvements; it is not encoded as a definitional equality. The paper does cite the authors' own Gar and Quam work for the alternating frontier/initial-pool schedule and for the corpus graphs, but those citations supply an independently published algorithmic substrate, not an unverified uniqueness theorem or an ansatz that is itself the conclusion. The one legitimate concern raised by the algorithm is that Algorithm 1 never explicitly specifies the final ordering of R1 when computing order-dependent nDCG@10; that is a reproducibility/correctness ambiguity to check against the released implementation, not a circularity, because the ordering is not defined in terms of the evaluation labels. Overall, the derivation chain from algorithm to measured effectiveness is falsifiable, externally benchmarked, and non-circular.

Assumptions & free parameters 3 free parameters · 4 assumptions · 1 invented entities

The central claim depends on three hand-chosen hyperparameters (w, b, k) and on four domain assumptions: the clustering hypothesis, the sufficiency of the precomputed graph, the validity of reciprocal-rank pseudo-scores, and the transferability of the alternating schedule from pointwise AR. None of these are fitted to the test qrels, but they are all load-bearing for the reported improvements.

free parameters (3)

window size w = 20
Set to the default used by the RankLLM/RankGPT sliding window implementation; all experiments use w=20, and it determines the number and size of LLM windows.
step size b = 10 (= w/2)
The paper assumes b = w/2; b controls how many documents are carried forward as survivors and how many are taken from the initial ranking or graph frontier each window.
graph depth k = 16
Number of neighbors per document in the corpus graph; re-used from Gar for MS MARCO v1 and built with 16 neighbors for v2. The ablation in Section 5.3 varies k, showing the method is sensitive to this choice.

assumptions (4)

domain assumption Clustering Hypothesis: documents similar to relevant documents are likely relevant to the same query.
SlideGar selects graph neighbors of the top-ranked documents in each window as candidate relevant documents; invoked in Section 2.2 and Algorithm 1 line 10.
domain assumption The precomputed corpus graph with k=16 nearest neighbors contains enough useful connections between initially retrieved and unretrieved relevant documents.
The method relies on graph lookups to supply documents missing from R0; if graph edges are missing or noisy, the bounded recall problem is not overcome. Introduced in Sections 3.1 and 4.1 and studied in the Section 5.3 ablation.
ad hoc to paper The reciprocal of the LLM's within-window rank is a usable pseudo-score for prioritizing graph neighbors.
Because listwise rankers give no numeric scores, Section 3.2 assigns reciprocal-of-rank pseudo-scores and uses them to order the graph frontier; this is an assumption, not a derived equivalence.
ad hoc to paper The top-b documents of each ranked window (L1) are reliable enough to carry forward as feedback, and the alternating R0/frontier schedule remains beneficial in the listwise setting.
Algorithm 1 carries L1 forward and alternates P between R0 and F following the Gar/Quam convention; the paper does not ablate this schedule against alternatives such as always preferring one source.

invented entities (1)

SlideGar (algorithmic procedure)
purpose: It is the proposed method for adaptive retrieval with listwise LLM rankers; not a new physical or theoretical entity.
SlideGar is an algorithm operating on existing corpus graphs and existing listwise rankers; it introduces no new particle, force, dimension, or conserved quantity.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Guiding Retrieval using LLM-based Listwise Rankers." pith.science (2026). https://pith.science/paper/2MTTCNKN

@misc{pith2026250109186,
  author       = {Pith},
  title        = {Pith review of: Guiding Retrieval using LLM-based Listwise Rankers},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/2MTTCNKN}},
  note         = {Machine review of arXiv:2501.09186}
}

read the original abstract

Large Language Models (LLMs) have shown strong promise as rerankers, especially in ``listwise'' settings where an LLM is prompted to rerank several search results at once. However, this ``cascading'' retrieve-and-rerank approach is limited by the bounded recall problem: relevant documents not retrieved initially are permanently excluded from the final ranking. Adaptive retrieval techniques address this problem, but do not work with listwise rerankers because they assume a document's score is computed independently from other documents. In this paper, we propose an adaptation of an existing adaptive retrieval method that supports the listwise setting and helps guide the retrieval process itself (thereby overcoming the bounded recall problem for LLM rerankers). Specifically, our proposed algorithm merges results both from the initial ranking and feedback documents provided by the most relevant documents seen up to that point. Through extensive experiments across diverse LLM rerankers, first stage retrievers, and feedback sources, we demonstrate that our method can improve nDCG@10 by up to 13.23% and recall by 28.02%--all while keeping the total number of LLM inferences constant and overheads due to the adaptive process minimal. The work opens the door to leveraging LLM-based search in settings where the initial pool of results is limited, e.g., by legacy systems, or by the cost of deploying a semantic first-stage.

Figures

Figures reproduced from arXiv: 2501.09186 by the authors.

**Figure 1.** The SlideGar algorithm visualized. LLM Ranker ranks the list of documents (window) and then SlideGar leverages LLM Ranker feedback and looks for the neighbors of d4 and d2, and carries both documents and their neighbors in the next window. The neighborhood documents are highlighted in green. The remaining documents, d1 and d3, are added to R1. the top b documents from F (denoted by L2) and prepare the window for th… view at source ↗

**Figure 2.** Effect of SlideGar on different ranking pipelines on TREC DL19 dataset when the number of graph neighbors k varies and window size is 20 and step size 10. rankers and do not consider the API-based rankers. We report the mean and variance over 5 consecutive runs. We conduct our experiment on the NVIDIA V100 GPU with 32G RAM. We observe that RankVicuna and RankZephyr take around 13797 ms and 14884 ms, respectively, to… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

39 extracted references · 13 canonical work pages

[1]

Computer Science Department Faculty Publication Series p

Abdul-Jaleel,N.,Allan,J.,Croft,W.B.,Diaz,F.,Larkey,L.,Li,X.,Smucker,M.D., Wade, C.: Umass at trec 2004: Novelty and hard. Computer Science Department Faculty Publication Series p. 189 (2004)

work page 2004
[2]

arXiv preprint arXiv:2306.15595 (2023)

Chen,S.,Wong,S.,Chen,L.,Tian,Y.:Extendingcontextwindowoflargelanguage models via positional interpolation. arXiv preprint arXiv:2306.15595 (2023)

arXiv 2023
[3]

Clavié,B.:rerankers:Alightweightpythonlibrarytounifyrankingmethods(2024), https://arxiv.org/abs/2408.17344

arXiv 2024
[4]

In: Ma, W., Nie, J., Baeza-Yates, R., Chua, T., Croft, W.B

Ding, S., Suel, T.: Faster top-k document retrieval using block-max indexes. In: Ma, W., Nie, J., Baeza-Yates, R., Chua, T., Croft, W.B. (eds.) Proceeding of the 34th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2011, Beijing, China, July 25-29, 2011. pp. 993–1002. ACM (2011). https://doi.org/10.1145/20099...

arXiv 2011
[5]

Jardine, N., van Rijsbergen, C.J.: The use of hierarchic clustering in informa- tion retrieval. Inf. Storage Retr. 7(5), 217–240 (1971). https://doi.org/10.1016/ 0020-0271(71)90051-9, https://doi.org/10.1016/0020-0271(71)90051-9

doi:10.1016/0020-0271(71)90051-9 1971
[6]

In: Proceedings of the 43rd International ACM SIGIR conference on research and development in Information Retrieval

Khattab, O., Zaharia, M.: Colbert: Efficient and effective passage search via con- textualized late interaction over bert. In: Proceedings of the 43rd International ACM SIGIR conference on research and development in Information Retrieval. pp. 39–48 (2020)

2020
[7]

In: Chen, H., Duh, W.E., Huang, H., Kato, M.P., Mothe, J., Poblete, B

Kulkarni, H., MacAvaney, S., Goharian, N., Frieder, O.: Lexically-accelerated dense retrieval. In: Chen, H., Duh, W.E., Huang, H., Kato, M.P., Mothe, J., Poblete, B. (eds.) Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2023, Taipei, Taiwan, July 23-27, 2023. pp. 152–162. ACM (2023). ...

arXiv 2023
[8]

In: Rogers, A., Calixto, I., Vulic, I., Saphra, N., Kassner, N., Camburu, O., Bansal, T., Shwartz, V

Lin, S., Yang, J., Lin, J.: In-batch negatives for knowledge distillation with tightly-coupled teachers for dense retrieval. In: Rogers, A., Calixto, I., Vulic, I., Saphra, N., Kassner, N., Camburu, O., Bansal, T., Shwartz, V. (eds.) Proceed- ings of the 6th Workshop on Representation Learning for NLP, RepL4NLP@ACL- IJCNLP 2021, Online, August 6, 2021. pp...

doi:10.18653/v1/2021.repl4nlp-1.17 2021

Show all 39 references

[9]

In: Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval

Ma, X., Wang, L., Yang, N., Wei, F., Lin, J.: Fine-tuning llama for multi-stage text retrieval. In: Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval. pp. 2421–2425 (2024)

2024
[10]

arXiv preprint arXiv:2305.02156 (2023)

Ma, X., Zhang, X., Pradeep, R., Lin, J.: Zero-shot listwise document reranking with a large language model. arXiv preprint arXiv:2305.02156 (2023)

2023 arXiv
[11]

In: Yang, G.H., Wang, H., Han, S., Hauff, C., Zuccon, G., Zhang, Y

MacAvaney, S., Tonellotto, N.: A reproducibility study of PLAID. In: Yang, G.H., Wang, H., Han, S., Hauff, C., Zuccon, G., Zhang, Y. (eds.) Proceed- ings of the 47th International ACM SIGIR Conference on Research and Devel- opment in Information Retrieval, SIGIR 2024, Washingt...

2024
[12]

MacAvaney, S., Tonellotto, N., Macdonald, C.: Adaptive re-ranking with a corpus graph.In:Hasan,M.A.,Xiong,L.(eds.)Proceedingsofthe31stACMInternational Conference on Information & Knowledge Management, Atlanta, GA, USA, Octo- ber 17-21, 2022. pp. 1491–1500. ACM (2022). https://...

2022
[13]

In: Demartini, G., Zuccon, G., Culpepper, J.S., Huang, Z., Tong, H

Macdonald, C., Tonellotto, N., MacAvaney, S., Ounis, I.: Pyterrier: Declarative ex- perimentation in python from BM25 to dense retrieval. In: Demartini, G., Zuccon, G., Culpepper, J.S., Huang, Z., Tong, H. (eds.) CIKM ’21: The 30th ACM Inter- national Conference on Information...

2021
[14]

arXiv preprint arXiv:2305.07477 (2023)

Mackie, I., Chatterjee, S., Dalton, J.: Generative and pseudo-relevant feedback for sparse, dense and learned sparse retrieval. arXiv preprint arXiv:2305.07477 (2023)

2023 arXiv
[15]

In: Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval

Mackie, I., Chatterjee, S., Dalton, J.: Generative relevance feedback with large language models. In: Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval. pp. 2026–2031 (2023)

2023
[16]

IEEE Trans

Malkov, Y.A., Yashunin, D.A.: Efficient and robust approximate nearest neigh- bor search using hierarchical navigable small world graphs. IEEE Trans. Pattern Anal. Mach. Intell.42(4), 824–836 (2020). https://doi.org/10.1109/TPAMI.2018. 2889473, https://doi.org/10.1109/TPAMI.20...

2020
[17]

In: Clancy, R., Ferro, N., Hauff, C., Lin, J., Sakai, T., Wu, Z.Z

Mallia, A., Siedlaczek, M., Mackenzie, J.M., Suel, T.: PISA: performant indexes and search for academia. In: Clancy, R., Ferro, N., Hauff, C., Lin, J., Sakai, T., Wu, Z.Z. (eds.) Proceedings of the Open-Source IR Replicability Challenge co- located with 42nd International ACM ...

2019
[18]

In: Efthimiadis, E.N., Dumais, S.T., Hawking, D., Järvelin, K

Matveeva, I., Burges, C., Burkard, T., Laucius, A., Wong, L.: High accuracy re- trieval with multiple nested ranker. In: Efthimiadis, E.N., Dumais, S.T., Hawking, D., Järvelin, K. (eds.) SIGIR 2006: Proceedings of the 29th Annual International ACM SIGIR Conference on Research ...

2006
[19]

In: Proceedings of the 28th annual international ACM SIGIR conference on Research and development in information retrieval

Metzler, D., Croft, W.B.: A markov random field model for term dependencies. In: Proceedings of the 28th annual international ACM SIGIR conference on Research and development in information retrieval. pp. 472–479 (2005)

2005
[20]

Metzler,D.,Croft,W.B.:Latentconceptexpansionusingmarkovrandomfields.In: Proceedings of the 30th annual international ACM SIGIR conference on Research and development in information retrieval. pp. 311–318 (2007)

2007
[21]

(eds.) Advances in Information Retrieval - 45th European Conference on Information Retrieval, ECIR 2023, Dublin, Ireland, April 2-6, 2023, Proceedings, Part III

Nguyen, T., MacAvaney, S., Yates, A.: A unified framework for learned sparse retrieval.In:Kamps,J.,Goeuriot,L.,Crestani,F.,Maistro,M.,Joho,H.,Davis,B., Gurrin, C., Kruschwitz, U., Caputo, A. (eds.) Advances in Information Retrieval - 45th European Conference on Information Ret...

2023 doi
[22]

In: Besold, T.R., Bordes, A., d’Avila Garcez, A.S., Wayne, G

Nguyen, T., Rosenberg, M., Song, X., Gao, J., Tiwary, S., Majumder, R., Deng, L.: MS MARCO: A human generated machine reading comprehension dataset. In: Besold, T.R., Bordes, A., d’Avila Garcez, A.S., Wayne, G. (eds.) Proceed- ings of the Workshop on Cognitive Computation: Int...

2016
[23]

In: Cohn, T., He, Y., Liu, Y

Nogueira, R.F., Jiang, Z., Pradeep, R., Lin, J.: Document ranking with a pre- trained sequence-to-sequence model. In: Cohn, T., He, Y., Liu, Y. (eds.) Findings of the Association for Computational Linguistics: EMNLP 2020, Online Event, 16-20 November 2020. Findings of ACL, vol...

2020 doi
[24]

https://openai.com (2024)

OpenAI: Openai api. https://openai.com (2024)

2024
[25]

arXiv preprint arXiv:2405.14589 (2024)

Parry, A., MacAvaney, S., Ganguly, D.: Top-down partitioning for efficient list-wise ranking. arXiv preprint arXiv:2405.14589 (2024)

2024 arXiv
[26]

arXiv preprint arXiv:2309.15088 (2023)

Pradeep, R., Sharifymoghaddam, S., Lin, J.: Rankvicuna: Zero-shot listwise document reranking with open-source large language models. arXiv preprint arXiv:2309.15088 (2023)

2023 arXiv
[27]

Pradeep, R., Sharifymoghaddam, S., Lin, J.: Rankzephyr: Effective and robust zero-shot listwise reranking is a breeze! arXiv preprint arXiv:2312.02724 (2023)

2023 arXiv
[28]

arXiv preprint arXiv:2306.17563 (2023)

Qin, Z., Jagerman, R., Hui, K., Zhuang, H., Wu, J., Yan, L., Shen, J., Liu, T., Liu, J., Metzler, D., et al.: Large language models are effective text rankers with pairwise ranking prompting. arXiv preprint arXiv:2306.17563 (2023)

2023 arXiv
[29]

arXiv preprint arXiv:2410.20286 (2024)

Rathee, M., MacAvaney, S., Anand, A.: Quam: Adaptive retrieval through query affinity modelling. arXiv preprint arXiv:2410.20286 (2024)

2024 arXiv
[30]

Journal of documentation 33(4), 294–304 (1977)

Robertson, S.E.: The probability ranking principle in ir. Journal of documentation 33(4), 294–304 (1977)

1977
[31]

Robertson, S.E., Zaragoza, H.: The probabilistic relevance framework: BM25 and beyond. Found. Trends Inf. Retr.3(4), 333–389 (2009). https://doi.org/10.1561/ 1500000019, https://doi.org/10.1561/1500000019

2009 doi
[32]

The SMART Retrieval System-Experiments in Automatic Document Processing/Prentice Hall (1971)

Rocchio, J.: Relevance feedback in information retrieval. The SMART Retrieval System-Experiments in Automatic Document Processing/Prentice Hall (1971)

1971
[33]

arXiv preprint arXiv:2406.14117 (2024)

Sun, S., Zhuang, S., Wang, S., Zuccon, G.: An investigation of prompt variations for zero-shot llm-based rankers. arXiv preprint arXiv:2406.14117 (2024)

2024 arXiv
[34]

arXiv preprint arXiv:2304.09542 (2023)

Sun, W., Yan, L., Ma, X., Wang, S., Ren, P., Chen, Z., Yin, D., Ren, Z.: Is chatgpt good at search? investigating large language models as re-ranking agents. arXiv preprint arXiv:2304.09542 (2023)

2023 arXiv
[35]

arXiv:2312.16098 (2023)

Tamber, M.S., Pradeep, R., Lin, J.: Scaling down, litting up: Efficient zero-shot listwise reranking with seq2seq encoder-decoder models. arXiv:2312.16098 (2023)

2023 arXiv
[36]

arXiv preprint arXiv:2403.15246 (2024)

Weller, O., Chang, B., MacAvaney, S., Lo, K., Cohan, A., Van Durme, B., Lawrie, D., Soldaini, L.: Followir: Evaluating and teaching information retrieval models to follow instructions. arXiv preprint arXiv:2403.15246 (2024)

2024 arXiv
[37]

In: Yang, G.H., Wang, H., Han, S., Hauff, C., Zuccon, G., Zhang, Y

Yang, Y., Carlson, P., He, S., Qiao, Y., Yang, T.: Cluster-based partial dense retrieval fused with sparse text retrieval. In: Yang, G.H., Wang, H., Han, S., Hauff, C., Zuccon, G., Zhang, Y. (eds.) Proceedings of the 47th International ACM SIGIR Conference on Research and Deve...

2024
[38]

In: Proceedings of the tenth international conference on Information and knowledge management

Zhai, C., Lafferty, J.: Model-based feedback in the language modeling approach to information retrieval. In: Proceedings of the tenth international conference on Information and knowledge management. pp. 403–410 (2001)

2001
[2019]

2409, pp

CEUR Workshop Proceedings, vol. 2409, pp. 50–56. CEUR-WS.org (2019), https://ceur-ws.org/Vol-2409/docker08.pdf

2019

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.

[1] [1]

Computer Science Department Faculty Publication Series p

Abdul-Jaleel,N.,Allan,J.,Croft,W.B.,Diaz,F.,Larkey,L.,Li,X.,Smucker,M.D., Wade, C.: Umass at trec 2004: Novelty and hard. Computer Science Department Faculty Publication Series p. 189 (2004)

work page 2004

[2] [2]

arXiv preprint arXiv:2306.15595 (2023)

Chen,S.,Wong,S.,Chen,L.,Tian,Y.:Extendingcontextwindowoflargelanguage models via positional interpolation. arXiv preprint arXiv:2306.15595 (2023)

arXiv 2023

[3] [3]

Clavié,B.:rerankers:Alightweightpythonlibrarytounifyrankingmethods(2024), https://arxiv.org/abs/2408.17344

arXiv 2024

[4] [4]

In: Ma, W., Nie, J., Baeza-Yates, R., Chua, T., Croft, W.B

Ding, S., Suel, T.: Faster top-k document retrieval using block-max indexes. In: Ma, W., Nie, J., Baeza-Yates, R., Chua, T., Croft, W.B. (eds.) Proceeding of the 34th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2011, Beijing, China, July 25-29, 2011. pp. 993–1002. ACM (2011). https://doi.org/10.1145/20099...

arXiv 2011

[5] [5]

Jardine, N., van Rijsbergen, C.J.: The use of hierarchic clustering in informa- tion retrieval. Inf. Storage Retr. 7(5), 217–240 (1971). https://doi.org/10.1016/ 0020-0271(71)90051-9, https://doi.org/10.1016/0020-0271(71)90051-9

doi:10.1016/0020-0271(71)90051-9 1971

[6] [6]

In: Proceedings of the 43rd International ACM SIGIR conference on research and development in Information Retrieval

Khattab, O., Zaharia, M.: Colbert: Efficient and effective passage search via con- textualized late interaction over bert. In: Proceedings of the 43rd International ACM SIGIR conference on research and development in Information Retrieval. pp. 39–48 (2020)

2020

[7] [7]

In: Chen, H., Duh, W.E., Huang, H., Kato, M.P., Mothe, J., Poblete, B

Kulkarni, H., MacAvaney, S., Goharian, N., Frieder, O.: Lexically-accelerated dense retrieval. In: Chen, H., Duh, W.E., Huang, H., Kato, M.P., Mothe, J., Poblete, B. (eds.) Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2023, Taipei, Taiwan, July 23-27, 2023. pp. 152–162. ACM (2023). ...

arXiv 2023

[8] [8]

In: Rogers, A., Calixto, I., Vulic, I., Saphra, N., Kassner, N., Camburu, O., Bansal, T., Shwartz, V

Lin, S., Yang, J., Lin, J.: In-batch negatives for knowledge distillation with tightly-coupled teachers for dense retrieval. In: Rogers, A., Calixto, I., Vulic, I., Saphra, N., Kassner, N., Camburu, O., Bansal, T., Shwartz, V. (eds.) Proceed- ings of the 6th Workshop on Representation Learning for NLP, RepL4NLP@ACL- IJCNLP 2021, Online, August 6, 2021. pp...

doi:10.18653/v1/2021.repl4nlp-1.17 2021