From 5c82f174de9427d09f88f5f26c5276c5356ba4ef Mon Sep 17 00:00:00 2001 From: lintool Date: Sun, 17 Sep 2023 08:48:06 -0400 Subject: [PATCH] Update 'Neural Hype' documentation. --- docs/experiments-forum2018.md | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/docs/experiments-forum2018.md b/docs/experiments-forum2018.md index bcce8f6e6c..210d33e5fe 100644 --- a/docs/experiments-forum2018.md +++ b/docs/experiments-forum2018.md @@ -12,7 +12,11 @@ The SIGIR Forum article references commit [`2c8cd7a`](https://github.com/castori The SIGIR 2019 paper contains experiments performed post upgrade. The Anserini upgrade to Lucene 8.0 at commit [`75e36f9`](https://github.com/castorini/anserini/commit/75e36f97f7037d1ceb20fa9c91582eac5e974131) (6/12/2019) broke the regression tests, which was later fixed at commit [`64bae9c`](https://github.com/castorini/anserini/commit/64bae9c8b87ad56bc8cf6ea0c5405eb2a82b3682) (7/3/2019). -This commit represents the latest state of the code and the results that can be currently reproduced. + +In September 2023, regression results were updated at commit [`6e148c6`](https://github.com/castorini/anserini/commit/6e148c6d22f78cb05ab0284d70701008b719ebc9) (2023/09/16). +This commit patched effectiveness differences arising from two main sources: (1) the upgrade to Lucene 9 at [`2725655`](https://github.com/castorini/anserini/commit/27256551e958f39495b04e89ef55de9d27f33414) (2022/08/02) and (2) a `fastutil` upgrade/bug fix at [#1975](https://github.com/castorini/anserini/pull/1975) that affected relevance feedback results. +To our knowledge, this commit represents the latest state of the code where the effectiveness encoded in our scripts can be successfully reproduced. + See summary in "History" section below. ## Expected Results @@ -108,6 +112,7 @@ This difference arises from rounding when averaging across the folds. The following documents commits that have altered effectiveness figures: ++ commit [`6e148c6`](https://github.com/castorini/anserini/commit/6e148c6d22f78cb05ab0284d70701008b719ebc9) (2023/09/16) - Regression experiments updated. + commit [`64bae9c`](https://github.com/castorini/anserini/commit/64bae9c8b87ad56bc8cf6ea0c5405eb2a82b3682) (7/3/2019) - Regression experiments here fixed. + commit [`75e36f9`](https://github.com/castorini/anserini/commit/75e36f97f7037d1ceb20fa9c91582eac5e974131) (6/12/2019) - Upgrade to Lucene 8.0 breaks regression experiments here. + commit [`407f308`](https://github.com/castorini/Anserini/commit/407f308cc543286e39701caf0acd1afab39dde2c) (1/2/2019) - Added results for axiomatic semantic term matching.