Sad I couldn't see your talk, @frommholz.org, as I was attending @ibnesayeed.bsky.social & @vphill.bsky.social excellent TAWAF tutorial on the last day of #jcdl2026
Do you have any slides or a recording you can share?
Do you have any slides or a recording you can share?
October 18, 2026 at 6:35 PM
Sad I couldn't see your talk, @frommholz.org, as I was attending @ibnesayeed.bsky.social & @vphill.bsky.social excellent TAWAF tutorial on the last day of #jcdl2026
Do you have any slides or a recording you can share?
Do you have any slides or a recording you can share?
Lamia Salsabil (@webscidl.bsky.social 🐸) begins her #jcdl2026 presentation, "End-to-End Contextualized URK Extraction from Scholarly Papers" saying there hasn't been a systematic study for robust methods for extracting URLs w/ their context.
October 15, 2026 at 6:57 PM
Lamia Salsabil (@webscidl.bsky.social 🐸) begins her #jcdl2026 presentation, "End-to-End Contextualized URK Extraction from Scholarly Papers" saying there hasn't been a systematic study for robust methods for extracting URLs w/ their context.
Leif closes his #jcdl2026 keynote with, "Libraries are the seed repository. Will we still be holding the seed when the next season comes?"
October 15, 2026 at 3:10 PM
Leif closes his #jcdl2026 keynote with, "Libraries are the seed repository. Will we still be holding the seed when the next season comes?"
At the Historical, Web, and Multimedia Archives session @ #jcdl2026, Bohan Jiang (@brandeisuniversity.bsky.social) descs his archival workflow with CLAMS to find a part of vintage video broadcasts.
Using agreement-based selection, they form the segments via 4 Pathways.
📄 doi.org/10.1145/3805...
Using agreement-based selection, they form the segments via 4 Pathways.
📄 doi.org/10.1145/3805...
October 15, 2026 at 4:18 PM
At the Historical, Web, and Multimedia Archives session @ #jcdl2026, Bohan Jiang (@brandeisuniversity.bsky.social) descs his archival workflow with CLAMS to find a part of vintage video broadcasts.
Using agreement-based selection, they form the segments via 4 Pathways.
📄 doi.org/10.1145/3805...
Using agreement-based selection, they form the segments via 4 Pathways.
📄 doi.org/10.1145/3805...
At the "Text and Language Processing" session at #jcdl2026, Shirin Shujaa from RMIT Univ. describes their controlled study of hallucination behavior under context compression.
They introduce Attention-Based Semantic Preservation, a method that retains tokens based on structural integrity.
They introduce Attention-Based Semantic Preservation, a method that retains tokens based on structural integrity.
October 15, 2026 at 6:37 PM
At the "Text and Language Processing" session at #jcdl2026, Shirin Shujaa from RMIT Univ. describes their controlled study of hallucination behavior under context compression.
They introduce Attention-Based Semantic Preservation, a method that retains tokens based on structural integrity.
They introduce Attention-Based Semantic Preservation, a method that retains tokens based on structural integrity.
To start the second day of #JCDL2026, Leif Azzopardi from @microsoft.com is presenting his keynote, "Design for Information Farmers".
Leif asks, "How can we reimagine designing for information farmers instead of information seekers?"
Leif asks, "How can we reimagine designing for information farmers instead of information seekers?"
October 15, 2026 at 2:25 PM
To start the second day of #JCDL2026, Leif Azzopardi from @microsoft.com is presenting his keynote, "Design for Information Farmers".
Leif asks, "How can we reimagine designing for information farmers instead of information seekers?"
Leif asks, "How can we reimagine designing for information farmers instead of information seekers?"
Boilerplate removal improves input. Neural and symbolic were developed to complement errors. #jcdl2026
Code: github.com/ylbmq/NSTS
Code: github.com/ylbmq/NSTS
October 15, 2026 at 4:52 PM
Boilerplate removal improves input. Neural and symbolic were developed to complement errors. #jcdl2026
Code: github.com/ylbmq/NSTS
Code: github.com/ylbmq/NSTS
Next, Ying-Chen Lin from Univ. Missouri talk about their work for identifying valid timestamps for RAG and #webarchiving #jcdl2026
Ads or related stories can have related stamps, too, which are then temporally aligned with the encapsulating page.
doi.org/10.1145/3805...
Ads or related stories can have related stamps, too, which are then temporally aligned with the encapsulating page.
doi.org/10.1145/3805...
October 15, 2026 at 4:37 PM
Next, Ying-Chen Lin from Univ. Missouri talk about their work for identifying valid timestamps for RAG and #webarchiving #jcdl2026
Ads or related stories can have related stamps, too, which are then temporally aligned with the encapsulating page.
doi.org/10.1145/3805...
Ads or related stories can have related stamps, too, which are then temporally aligned with the encapsulating page.
doi.org/10.1145/3805...
Their baseline and evaluation first required that the URL matching exactly after normalizations.
The ref model oerformed the worst vs VisURLX variants.
#jcdl2026
The ref model oerformed the worst vs VisURLX variants.
#jcdl2026
October 15, 2026 at 7:10 PM
Their baseline and evaluation first required that the URL matching exactly after normalizations.
The ref model oerformed the worst vs VisURLX variants.
#jcdl2026
The ref model oerformed the worst vs VisURLX variants.
#jcdl2026
Information farming library isn't categorically different. It can tell when knowledge artifacts become trustworthy.
Specific aim isn't individual learning but the cultivation of public ideas. Pruning and weeding become cheaper.
The cultivation and knowledge now is much more reliable.
#jcdl2026
Specific aim isn't individual learning but the cultivation of public ideas. Pruning and weeding become cheaper.
The cultivation and knowledge now is much more reliable.
#jcdl2026
October 15, 2026 at 2:49 PM
Information farming library isn't categorically different. It can tell when knowledge artifacts become trustworthy.
Specific aim isn't individual learning but the cultivation of public ideas. Pruning and weeding become cheaper.
The cultivation and knowledge now is much more reliable.
#jcdl2026
Specific aim isn't individual learning but the cultivation of public ideas. Pruning and weeding become cheaper.
The cultivation and knowledge now is much more reliable.
#jcdl2026
In a survey, over 90% of attacks were attributed to aggressive bots. Humans vs bot is not the real design problem.
Are the bots doing anything for the user or just bull training, exploiting? The new user is the agent.
"I probably have more bot users on Twitter than humans."
#JCDL2026
Are the bots doing anything for the user or just bull training, exploiting? The new user is the agent.
"I probably have more bot users on Twitter than humans."
#JCDL2026
October 15, 2026 at 2:53 PM
In a survey, over 90% of attacks were attributed to aggressive bots. Humans vs bot is not the real design problem.
Are the bots doing anything for the user or just bull training, exploiting? The new user is the agent.
"I probably have more bot users on Twitter than humans."
#JCDL2026
Are the bots doing anything for the user or just bull training, exploiting? The new user is the agent.
"I probably have more bot users on Twitter than humans."
#JCDL2026
Their results imply that it's cheaper to peform the process using a prompt. It's both less expensive and more accurate.
VusURLX (w/ GPT 5-mini) had the highest F1 score for extracting URLs and retaining context.
#jcdl2026
VusURLX (w/ GPT 5-mini) had the highest F1 score for extracting URLs and retaining context.
#jcdl2026
October 15, 2026 at 7:14 PM
Their results imply that it's cheaper to peform the process using a prompt. It's both less expensive and more accurate.
VusURLX (w/ GPT 5-mini) had the highest F1 score for extracting URLs and retaining context.
#jcdl2026
VusURLX (w/ GPT 5-mini) had the highest F1 score for extracting URLs and retaining context.
#jcdl2026
Different extraction methods identify and extract date candidates including consideration of confidence and date conflicts from their methods.
If confidence, they select a neural timestamp. If low, they select the symbolic timestamp.
#jcdl2026
If confidence, they select a neural timestamp. If low, they select the symbolic timestamp.
#jcdl2026
October 15, 2026 at 4:42 PM
Different extraction methods identify and extract date candidates including consideration of confidence and date conflicts from their methods.
If confidence, they select a neural timestamp. If low, they select the symbolic timestamp.
#jcdl2026
If confidence, they select a neural timestamp. If low, they select the symbolic timestamp.
#jcdl2026
It's not clicks and downloads, it's whether the knowledge was sound.
Doc isn't the endpoint anymore. Library's competition isn't another repo, it's that the agent is the users' primary interface.
#jcdl2026
Doc isn't the endpoint anymore. Library's competition isn't another repo, it's that the agent is the users' primary interface.
#jcdl2026
October 15, 2026 at 3:05 PM
It's not clicks and downloads, it's whether the knowledge was sound.
Doc isn't the endpoint anymore. Library's competition isn't another repo, it's that the agent is the users' primary interface.
#jcdl2026
Doc isn't the endpoint anymore. Library's competition isn't another repo, it's that the agent is the users' primary interface.
#jcdl2026
Normally a good segment is 15-16 sec. long. With the same transcript, all pipelines summarize equally well. #jcdl2026
Future work: indexing summaries and merging and splitting rules that steer segments toward rundown-item length.
Future work: indexing summaries and merging and splitting rules that steer segments toward rundown-item length.
October 15, 2026 at 4:24 PM
Normally a good segment is 15-16 sec. long. With the same transcript, all pipelines summarize equally well. #jcdl2026
Future work: indexing summaries and merging and splitting rules that steer segments toward rundown-item length.
Future work: indexing summaries and merging and splitting rules that steer segments toward rundown-item length.
Does anyone remember Gopher? ✋
These changes in tech linked significant passages and activities across documents. The TREC program looked to access to large amounts (~ 1 GB at the time) of machine-readable docs.
#JCDL2026
These changes in tech linked significant passages and activities across documents. The TREC program looked to access to large amounts (~ 1 GB at the time) of machine-readable docs.
#JCDL2026
October 14, 2026 at 7:38 PM
Does anyone remember Gopher? ✋
These changes in tech linked significant passages and activities across documents. The TREC program looked to access to large amounts (~ 1 GB at the time) of machine-readable docs.
#JCDL2026
These changes in tech linked significant passages and activities across documents. The TREC program looked to access to large amounts (~ 1 GB at the time) of machine-readable docs.
#JCDL2026
Their process for associating context with URLs in scholarly docs:
1. Segment to paper components.
2. ID bottom of page
3. Get bib entries
4. ID boilerplate
Their annotation was 97.4% reliable every diagreement was reviewed and any confirmed errors were corrected.
#jcdl2026
1. Segment to paper components.
2. ID bottom of page
3. Get bib entries
4. ID boilerplate
Their annotation was 97.4% reliable every diagreement was reviewed and any confirmed errors were corrected.
#jcdl2026
October 15, 2026 at 7:05 PM
Their process for associating context with URLs in scholarly docs:
1. Segment to paper components.
2. ID bottom of page
3. Get bib entries
4. ID boilerplate
Their annotation was 97.4% reliable every diagreement was reviewed and any confirmed errors were corrected.
#jcdl2026
1. Segment to paper components.
2. ID bottom of page
3. Get bib entries
4. ID boilerplate
Their annotation was 97.4% reliable every diagreement was reviewed and any confirmed errors were corrected.
#jcdl2026
October 14, 2026 at 7:12 PM
"If a URL is in a footnote, you first have to restore it to the sentence where it is used."
There are benchmarks for scholarly URL identif.
To address the gap, they created a benchmark, created an LLM, compared compression, and evaluated the extraction process.
#jcdl2026
There are benchmarks for scholarly URL identif.
To address the gap, they created a benchmark, created an LLM, compared compression, and evaluated the extraction process.
#jcdl2026
October 15, 2026 at 7:01 PM
"If a URL is in a footnote, you first have to restore it to the sentence where it is used."
There are benchmarks for scholarly URL identif.
To address the gap, they created a benchmark, created an LLM, compared compression, and evaluated the extraction process.
#jcdl2026
There are benchmarks for scholarly URL identif.
To address the gap, they created a benchmark, created an LLM, compared compression, and evaluated the extraction process.
#jcdl2026
At the first day of #JCDL2026, Nicholas Belkin (@rutgersuniversity.bsky.social) starts his keynote "Digital Libraries and Their Technologies: A Look Back and a Peek Forward" by reminiscing on his involvement in organizing the conference 25 years ago under the auspices of @tamu.bsky.social.
October 14, 2026 at 6:45 PM
At the first day of #JCDL2026, Nicholas Belkin (@rutgersuniversity.bsky.social) starts his keynote "Digital Libraries and Their Technologies: A Look Back and a Peek Forward" by reminiscing on his involvement in organizing the conference 25 years ago under the auspices of @tamu.bsky.social.
The problem isn't about price, it's about consent. It's both a danger and leverage. Every usage metric will tell that the valuable library is being unused.
Agents need curated corpora to give them provenance to trust what they've produced.
#jcdl2026
Agents need curated corpora to give them provenance to trust what they've produced.
#jcdl2026
October 15, 2026 at 2:58 PM
The problem isn't about price, it's about consent. It's both a danger and leverage. Every usage metric will tell that the valuable library is being unused.
Agents need curated corpora to give them provenance to trust what they've produced.
#jcdl2026
Agents need curated corpora to give them provenance to trust what they've produced.
#jcdl2026
Fine-tuning helps ~7x but remaining gal is coverage, not ordering. Their pipeline has 3 stages, each targeting coverage, no extra labels.
Shuffling the tag order teaches the set. Voting gives a modenfor precision-recall. Many "misses" look like valid subjects that the record left out.
#jcdl2026
Shuffling the tag order teaches the set. Voting gives a modenfor precision-recall. Many "misses" look like valid subjects that the record left out.
#jcdl2026
October 15, 2026 at 5:07 PM
Fine-tuning helps ~7x but remaining gal is coverage, not ordering. Their pipeline has 3 stages, each targeting coverage, no extra labels.
Shuffling the tag order teaches the set. Voting gives a modenfor precision-recall. Many "misses" look like valid subjects that the record left out.
#jcdl2026
Shuffling the tag order teaches the set. Voting gives a modenfor precision-recall. Many "misses" look like valid subjects that the record left out.
#jcdl2026