LLMsForDisinformationAnalysis/agent at 8c0921057bfe8b28c78a051c92cfb804824a2228 - LLMsForDisinformationAnalysis - Gitea on Jillbox

jill/LLMsForDisinformationAnalysis

Files

T

History

William Jeynes 8c0921057b start on work to calculate % if valid URLS

2026-04-04 18:52:47 +01:00

..

Refactor calculating score. Add sort node for vanity

2026-02-12 23:46:00 +00:00

Add more robust parsing of LLM JSON output

2026-03-27 11:09:59 +00:00

add date to returned data

2026-03-25 22:37:14 +00:00

Fix errors seen during longer runs: selenium exceptions, insecure certificates, recusrsion limit exceeded, BM25 document corpus too small

2026-03-26 12:22:13 +00:00

Add file logging for errors. Add exponential backoff retry to web search. On failed web search, do not crash pipeline, return placeholder text to loanguage model

2026-02-24 13:05:35 +00:00

.env.example

FEAT: implement temp version of main tooling feedback loop

2026-02-09 20:25:36 +00:00

.gitignore

Move all data to own folder. Add run shell script. Experiment (unsuccessfully so far) with example retreival

2026-02-16 22:42:13 +00:00

agent.ts

Implement ensemble into final model structure

2026-03-24 19:07:24 +00:00

langgraph.json

Add re-ranker mode to support re-ranking experiments, hopefully we can reduce the loss

2026-03-06 17:27:09 +00:00

package-lock.json

Add more robust parsing of LLM JSON output

2026-03-27 11:09:59 +00:00

package.json

Add more robust parsing of LLM JSON output

2026-03-27 11:09:59 +00:00

README.md

start on work to calculate % if valid URLS

2026-04-04 18:52:47 +01:00

state.ts

add date to returned data

2026-03-25 22:37:14 +00:00

verify.ts

Implement ensemble into final model structure

2026-03-24 19:07:24 +00:00

README.md

Refining the agent output

Experiments modifying pipeline

Model	% Correct	% Change
BASELINE	33	0
Improv Prompt	39.96	0.21
Add Examples	44.67	0.35
Date	45.51	0.38
Chain of Thought	43.38	0.31
Self-Critique	44.36	0.34

Experiments with different model types:

Model	% Correct	% Change
gpt-5-mini	33	0
gpt-5.4-mini	32.4	-0.02
llama3.1:8b-instruct-q4_K_M	?	?
qwen3.5:9b	0	-100

%age correct URLS