FullFact health-misinfo-shared issues

FullFact / health-misinfo-shared

Raphael health misinformation project, shared by Full Fact and Google

MIT License

0 stars 0 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

(placeholder) Distil Raphael's Gemini model into a small, standalone model

#153 dcorney opened 1 day ago
0
(placeholder) Split this repo into two projects

#152 dcorney opened 1 day ago
0
Simultaneous submissions cause slow reponses

#151 ff-dh opened 1 day ago
2
feat: only show current user’s video history

#150 andylolz closed 1 day ago
0
Make sure different users’ video requests are only available to them

#149 andylolz closed 1 day ago
5
refactor: assume we have exactly one candidate

#148 andylolz closed 2 days ago
1
Run black, flake8 and isort over everything

#147 andylolz opened 3 days ago
0
Create a distilled version of Raphael's LLM

#146 dcorney opened 3 days ago
0
LLM returns exactly one candidate. We should modify the code accordingly

#145 andylolz closed 2 days ago
2
140 add format checker

#144 c-j-johnston closed 3 days ago
0
*** problem handling output? *** Content has no parts.

#143 andylolz opened 1 week ago
2
feat: don't limit number of claims per chunk

#142 c-j-johnston closed 1 week ago
0
fix: handle missing keys in LLM response

#141 andylolz closed 1 week ago
1
Repeat claim extraction when LLM output is wrong

#140 c-j-johnston closed 3 days ago
3
feat: parse output jsons using existing function

#139 c-j-johnston closed 1 week ago
0
Use json parse function in inference

#138 c-j-johnston closed 1 week ago
0
fix: add missing data file

#137 c-j-johnston closed 1 week ago
0
add missing data file

#136 c-j-johnston closed 1 week ago
0
fix: add claim summaries again for app

#135 c-j-johnston closed 2 weeks ago
0
fix: only call google.auth.default() from inside a function

#134 andylolz closed 2 weeks ago
2
Always call google.auth.default from inside a function

#133 andylolz closed 2 weeks ago
0
131 remove evaluationpy

#132 dearden closed 2 weeks ago
0
Remove evaluation.py

#131 dearden closed 2 weeks ago
2
Deployment Process for Raphael Image

#130 ff-dh opened 2 weeks ago
0
fix: check end_idx before using it

#129 andylolz closed 2 weeks ago
2
chore: update readme for new users handling

#128 ff-dh closed 3 weeks ago
0
91 improve generation of labels for in context learning

#127 c-j-johnston closed 2 weeks ago
0
fix: remove text from labels

#126 andylolz closed 2 weeks ago
0
Error in refine_offsets causes transcript processing to fail to complete

#125 andylolz closed 2 weeks ago
0
Replace match label words with just colours

#124 scoltman closed 2 weeks ago
0
fix: use user_id in runs & rename disable route

#123 ff-dh closed 3 weeks ago
0
Consolidate json parsing into one implementation

#122 dearden opened 4 weeks ago
0
feat: user registration routes

#121 ff-dh closed 3 weeks ago
2
Video transcripts should be punctuated

#120 JamesMcMinn closed 2 weeks ago
2
Add user creation, deletion and RBAC for transcript creation

#119 ff-dh closed 3 weeks ago
0
Allow for user feedback on inferred claims

#118 scoltman opened 4 weeks ago
0
116 improve json parsing of generated output

#117 dearden closed 4 weeks ago
0
Improve json parsing of generated output

#116 dearden closed 4 weeks ago
0
fix: update captions regex

#115 andylolz closed 1 month ago
0
Optimise chunk size

#114 dcorney closed 1 week ago
1
Captions aren’t extracted for some videos

#113 andylolz closed 1 month ago
2
Evaluation assertions

#112 dearden closed 4 weeks ago
1
UI improvements

#111 scoltman opened 1 month ago
1
feat: show transcript alongside claims

#110 andylolz closed 1 month ago
0
fix: limit history to 20 items

#109 andylolz closed 1 month ago
0
fix: pass raw transcript text to LLM

#108 andylolz closed 1 month ago
0
Pass plaintext chunk to the LLM (instead of a stringified dict)

#107 andylolz closed 1 month ago
1
Sometimes the YouTube embed doesn’t load

#106 andylolz opened 1 month ago
0
Add a test for checking the format of the LLM output

#105 dearden closed 4 weeks ago
0
Indicate that a video analysis is incomplete

#104 andylolz closed 1 month ago
1