Proceedings of the 30th annual international ACM SIGIR conference on Research and development in information retrieval
Bibliographic record
Abstract
Welcome to the 30th year of SIGIR, the Annual International ACM SIGIR Conference on Research and Development in Information Retrieval. The growth in SIGIR over the past few years has been remarkable. SIGIR 2005 saw a record 368 full paper submissions; SIGIR 2006 again set a record with 399 submissions. In our planning for 2007 we anticipated between 350 and 400 papers. Instead, we were pleasantly surprised, but somewhat overwhelmed, when we received 490 submissions, an increase of nearly 23% over 2006. From these submissions we were able to accept 84 papers (17.5%). These contributions are drawn from the breadth of IR research, including indexing, efficiency, evaluation, formal models, machine learning, classification, and user studies. Many papers are devoted to specialized topics such as question answering, multi-media retrieval, Web search and spam. Along with these full papers, we were able to accept 105 posters, 18 demonstrations, 9 tutorials and 9 workshops. In addition, 9 PhD candidates were selected to participate in our doctoral consortium. Again this year, the selection of full papers was dependent on a two-tiered reviewing process. The PC Chairs and 33 Senior PC members, nominated nearly 313 primary reviewers. Each reviewer was assigned between 3 and 8 papers by the PC chairs in accordance to reviewers' stated subject expertise and each paper was allocated three reviewers. In cases where there was a wide range of scores or incomplete information, several primary reviewers helped us out with a number of additional reviews. The role of the Senior PC members was to oversee the review process, by resolving disagreements between reviewers and producing a meta-review for each paper. These meta-reviews served as the basis for discussion at the Programme Committee meeting. Each Senior PC member was responsible for 13 to 18 papers. Senior PC members were selected for their subject expertise in the different topic areas and attention was also paid to geographic representation and PC Committee experience. All the reviewing was double blind with the identity of authors being released only after the selection of papers was completed. We thank the members of the Senior PC for their hard work in helping us cope with the unexpected number of submissions. Similarly thorough processes were followed for the selection of posters, demonstrations, tutorials, and workshops, as well as for the selection of participants in the doctoral consortium. We are grateful for the efforts of the various chairs who managed the selection of these contributions: Gianni Amati, Chris Buckley, Thomas Hofmann, Liz Liddy, Josiane Mothe, Thomas Roelleke, Mark Sanderson, and ChengXiang Zhai. We thank Edwin van Huis, our keynote speaker, for agreeing to share his ideas with us. We thank the SIGIR executive committee for their willingness to answer our many questions quickly and carefully. We thank TNO, who hosted the Program Committee meeting in Delft. Finally, we thank Wessel Kraaij and Arjen P. de Vries, the conference General Chairs for their tremendous efforts to make this conference a success. In early April we were saddened to learn of the passing of Karen Sparck Jones, one of the great pioneers of information retrieval. We are pleased to host her Athena Award lecture, and we dedicate these proceedings to her memory.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.010 | 0.014 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.003 | 0.002 |
| Bibliometrics | 0.004 | 0.002 |
| Science and technology studies | 0.002 | 0.002 |
| Scholarly communication | 0.010 | 0.009 |
| Open science | 0.003 | 0.004 |
| Research integrity | 0.003 | 0.006 |
| Insufficient payload (model declined to judge) | 0.136 | 0.113 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".