This page describes experiments, integrated into Anserini's regression testing framework, for the TREC 2020 Deep Learning Track (Document Ranking Task) on the MS MARCO document collection using relevance judgments from NIST.
Note that the NIST relevance judgments provide far more relevant documents per topic, unlike the "sparse" judgments provided by Microsoft (these are sometimes called "dense" judgments to emphasize this contrast). For additional instructions on working with MS MARCO document collection, refer to this page.
Note that there are four different regression conditions for this task, and this page describes the following:
- Indexing Condition: each MS MARCO document is treated as a unit of indexing
- Expansion Condition: none
All four conditions are described in detail here, in the context of doc2query-T5.
The exact configurations for these regressions are stored in this YAML file. Note that this page is automatically generated from this template as part of Anserini's regression pipeline, so do not modify this page directly; modify the template instead.
Note that in November 2021 we discovered issues in our regression tests, documented here. As a result, we have had to rebuild all our regressions from the raw corpus. These new versions yield end-to-end scores that are slightly different, so if numbers reported in a paper do not exactly match the numbers here, this may be the reason.
From one of our Waterloo servers (e.g., orca), the following command will perform the complete regression, end to end:
python src/main/python/run_regression.py --index --verify --search --regression dl20-doc
Typical indexing command:
target/appassembler/bin/IndexCollection \
-collection JsonCollection \
-input /path/to/msmacro-doc \
-index indexes/lucene-index.msmarco-doc/ \
-generator DefaultLuceneDocumentGenerator \
-threads 7 -storePositions -storeDocvectors -storeRaw \
>& logs/log.msmacro-doc &
The directory /path/to/msmarco-doc/ should be a directory containing the document corpus in Anserini's jsonl format.
See this page for how to prepare the corpus.
For additional details, see explanation of common indexing options.
Topics and qrels are stored in src/main/resources/topics-and-qrels/.
The regression experiments here evaluate on the 45 topics for which NIST has provided judgments as part of the TREC 2020 Deep Learning Track.
The original data can be found here.
After indexing has completed, you should be able to perform retrieval as follows:
target/appassembler/bin/SearchCollection \
-index indexes/lucene-index.msmarco-doc/ \
-topics src/main/resources/topics-and-qrels/topics.dl20.txt \
-topicreader TsvInt \
-output runs/run.msmacro-doc.bm25-default.topics.dl20.txt \
-bm25 &
target/appassembler/bin/SearchCollection \
-index indexes/lucene-index.msmarco-doc/ \
-topics src/main/resources/topics-and-qrels/topics.dl20.txt \
-topicreader TsvInt \
-output runs/run.msmacro-doc.bm25-default+rm3.topics.dl20.txt \
-bm25 -rm3 &
target/appassembler/bin/SearchCollection \
-index indexes/lucene-index.msmarco-doc/ \
-topics src/main/resources/topics-and-qrels/topics.dl20.txt \
-topicreader TsvInt \
-output runs/run.msmacro-doc.bm25-tuned.topics.dl20.txt \
-bm25 -bm25.k1 3.44 -bm25.b 0.87 &
target/appassembler/bin/SearchCollection \
-index indexes/lucene-index.msmarco-doc/ \
-topics src/main/resources/topics-and-qrels/topics.dl20.txt \
-topicreader TsvInt \
-output runs/run.msmacro-doc.bm25-tuned+rm3.topics.dl20.txt \
-bm25 -bm25.k1 3.44 -bm25.b 0.87 -rm3 &
target/appassembler/bin/SearchCollection \
-index indexes/lucene-index.msmarco-doc/ \
-topics src/main/resources/topics-and-qrels/topics.dl20.txt \
-topicreader TsvInt \
-output runs/run.msmacro-doc.bm25-tuned2.topics.dl20.txt \
-bm25 -bm25.k1 4.46 -bm25.b 0.82 &
target/appassembler/bin/SearchCollection \
-index indexes/lucene-index.msmarco-doc/ \
-topics src/main/resources/topics-and-qrels/topics.dl20.txt \
-topicreader TsvInt \
-output runs/run.msmacro-doc.bm25-tuned2+rm3.topics.dl20.txt \
-bm25 -bm25.k1 4.46 -bm25.b 0.82 -rm3 &
Evaluation can be performed using trec_eval:
tools/eval/trec_eval.9.0.4/trec_eval -c -M 100 -m map src/main/resources/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmacro-doc.bm25-default.topics.dl20.txt
tools/eval/trec_eval.9.0.4/trec_eval -c -m ndcg_cut.10 src/main/resources/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmacro-doc.bm25-default.topics.dl20.txt
tools/eval/trec_eval.9.0.4/trec_eval -c -m recall.100 src/main/resources/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmacro-doc.bm25-default.topics.dl20.txt
tools/eval/trec_eval.9.0.4/trec_eval -c -m recall.1000 src/main/resources/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmacro-doc.bm25-default.topics.dl20.txt
tools/eval/trec_eval.9.0.4/trec_eval -c -M 100 -m map src/main/resources/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmacro-doc.bm25-default+rm3.topics.dl20.txt
tools/eval/trec_eval.9.0.4/trec_eval -c -m ndcg_cut.10 src/main/resources/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmacro-doc.bm25-default+rm3.topics.dl20.txt
tools/eval/trec_eval.9.0.4/trec_eval -c -m recall.100 src/main/resources/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmacro-doc.bm25-default+rm3.topics.dl20.txt
tools/eval/trec_eval.9.0.4/trec_eval -c -m recall.1000 src/main/resources/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmacro-doc.bm25-default+rm3.topics.dl20.txt
tools/eval/trec_eval.9.0.4/trec_eval -c -M 100 -m map src/main/resources/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmacro-doc.bm25-tuned.topics.dl20.txt
tools/eval/trec_eval.9.0.4/trec_eval -c -m ndcg_cut.10 src/main/resources/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmacro-doc.bm25-tuned.topics.dl20.txt
tools/eval/trec_eval.9.0.4/trec_eval -c -m recall.100 src/main/resources/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmacro-doc.bm25-tuned.topics.dl20.txt
tools/eval/trec_eval.9.0.4/trec_eval -c -m recall.1000 src/main/resources/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmacro-doc.bm25-tuned.topics.dl20.txt
tools/eval/trec_eval.9.0.4/trec_eval -c -M 100 -m map src/main/resources/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmacro-doc.bm25-tuned+rm3.topics.dl20.txt
tools/eval/trec_eval.9.0.4/trec_eval -c -m ndcg_cut.10 src/main/resources/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmacro-doc.bm25-tuned+rm3.topics.dl20.txt
tools/eval/trec_eval.9.0.4/trec_eval -c -m recall.100 src/main/resources/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmacro-doc.bm25-tuned+rm3.topics.dl20.txt
tools/eval/trec_eval.9.0.4/trec_eval -c -m recall.1000 src/main/resources/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmacro-doc.bm25-tuned+rm3.topics.dl20.txt
tools/eval/trec_eval.9.0.4/trec_eval -c -M 100 -m map src/main/resources/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmacro-doc.bm25-tuned2.topics.dl20.txt
tools/eval/trec_eval.9.0.4/trec_eval -c -m ndcg_cut.10 src/main/resources/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmacro-doc.bm25-tuned2.topics.dl20.txt
tools/eval/trec_eval.9.0.4/trec_eval -c -m recall.100 src/main/resources/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmacro-doc.bm25-tuned2.topics.dl20.txt
tools/eval/trec_eval.9.0.4/trec_eval -c -m recall.1000 src/main/resources/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmacro-doc.bm25-tuned2.topics.dl20.txt
tools/eval/trec_eval.9.0.4/trec_eval -c -M 100 -m map src/main/resources/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmacro-doc.bm25-tuned2+rm3.topics.dl20.txt
tools/eval/trec_eval.9.0.4/trec_eval -c -m ndcg_cut.10 src/main/resources/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmacro-doc.bm25-tuned2+rm3.topics.dl20.txt
tools/eval/trec_eval.9.0.4/trec_eval -c -m recall.100 src/main/resources/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmacro-doc.bm25-tuned2+rm3.topics.dl20.txt
tools/eval/trec_eval.9.0.4/trec_eval -c -m recall.1000 src/main/resources/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmacro-doc.bm25-tuned2+rm3.topics.dl20.txt
With the above commands, you should be able to reproduce the following results:
| AP@100 | BM25 (default) | +RM3 | BM25 (tuned) | +RM3 | BM25 (tuned2) | +RM3 |
|---|---|---|---|---|---|---|
| DL20 (Doc) | 0.3793 | 0.4014 | 0.3631 | 0.3592 | 0.3581 | 0.3619 |
| nDCG@10 | BM25 (default) | +RM3 | BM25 (tuned) | +RM3 | BM25 (tuned2) | +RM3 |
|---|---|---|---|---|---|---|
| DL20 (Doc) | 0.5286 | 0.5225 | 0.5070 | 0.5124 | 0.5061 | 0.5238 |
| R@100 | BM25 (default) | +RM3 | BM25 (tuned) | +RM3 | BM25 (tuned2) | +RM3 |
|---|---|---|---|---|---|---|
| DL20 (Doc) | 0.6110 | 0.6414 | 0.5935 | 0.5977 | 0.5860 | 0.5995 |
| R@1000 | BM25 (default) | +RM3 | BM25 (tuned) | +RM3 | BM25 (tuned2) | +RM3 |
|---|---|---|---|---|---|---|
| DL20 (Doc) | 0.8085 | 0.8257 | 0.7876 | 0.8116 | 0.7776 | 0.8180 |
Explanation of settings:
- The setting "default" refers the default BM25 settings of
k1=0.9,b=0.4. - The setting "tuned" refers to
k1=3.44,b=0.87, tuned on 2019/06 and used for TREC 2019 Deep Learning Track baseline runs. - The setting "tuned2" refers to
k1=4.46,b=0.82, tuned using the MS MARCO document sparse judgments to optimize for recall@100 (i.e., for first-stage retrieval) on 2019/12; see this page additional details.
Settings tuned on the MS MARCO document sparse judgments may not work well on the TREC dense judgments.
Note that in the official evaluation for document ranking, all runs were truncated to top-100 hits per query (whereas all top-1000 hits per query were retained for passage ranking). Thus, average precision is computed to depth 100 (i.e., AP@100); nDCG@10 remains unaffected. Remember that we keep qrels of all relevance grades, unlike the case for passage ranking, where relevance grade 1 needs to be discarded when computing certain metrics. Here, we retrieve 1000 hits per query, but measure AP at cutoff 100 (e.g., AP@100). Thus, the experimental results reported here are directly comparable to the results reported in the track overview paper.
Some of these regressions correspond to official TREC 2020 Deep Learning Track submissions by team anserini:
d_bm25= BM25 (default),k1=0.9,b=0.4d_bm25rm3= BM25 (default) + RM3,k1=0.9,b=0.4
Note that #1721 slightly change the results, since we corrected underlying issues with data preparation.