Repository navigation

Benchmark for Chat History Compression Algorithms in Agentic Systems with Function Calling - #13

Open

mikhail-kharlamov wants to merge 40 commits into

emnigma:mainfrom

mikhail-kharlamov:benchmarking-first-stage

Commits on Nov 9, 2025
- feat: added launch of new evaluation
  michaelharlam
  committed
  488fe6b
- refactoring 1
  michaelharlam
  committed
  6c0bd10
- refactoring 2
  michaelharlam
  committed
  66fc521
- refactoring 3
  michaelharlam
  committed
  0944ee5
- feat: evaluating is ready for launching
  michaelharlam
  committed
  abf4bb7
Commits on Nov 12, 2025
- feat: added f1 score tool evaluator
  michaelharlam
  committed
  dace504
- feat: launch f1
  michaelharlam
  committed
  d0c10b0
Commits on Nov 15, 2025
- feat: added statistics
  michaelharlam
  committed
  862a964
Commits on Nov 16, 2025
- feat: added working statistics calculating
  michaelharlam
  committed
  3e05acb
- feat: jsonl -> json in memory
  michaelharlam
  committed
  df80de8
Commits on Nov 17, 2025
- feat: refactoring 1
  michaelharlam
  committed
  e4b8544
- feat: added parsing new data type
  michaelharlam
  committed
  68953a5
Commits on Nov 18, 2025
- feat: added color logs, shuffled launch
  michaelharlam
  committed
  afe523e
Commits on Dec 25, 2025
- benchmark + new algorithms
  michaelharlam
  committed
  7e263fb
Commits on Dec 27, 2025
- feat: rewrite from string prompt to list of messages (only baseline dialogue)
  michaelharlam
  committed
  093b202
Commits on Dec 28, 2025
- feat: added DialogueWithWeights
  michaelharlam
  committed
  8d2c76e
Commits on Jan 6, 2026
- fix: style
  michaelharlam
  committed
  e54a59e
Commits on Jan 7, 2026
- feat: added langchain messages to memory algs. also changes by mypy, by ruff and in tests
  michaelharlam
  committed
  50b3e80
- fix: ruff and mypy
  michaelharlam
  committed
  9e371d8
- fix: some important constants move to .env
  michaelharlam
  committed
  2c75ff8
- fix: mypy & ruff
  michaelharlam
  committed
  186d0b8
- fix: pytest version
  michaelharlam
  committed
  1d5223c
Commits on Jan 24, 2026
- feat: added separation into a main model and the model for memory; added feature for using local models
  michaelharlam
  committed
  50f6bbd
- fix: deleted extra file
  michaelharlam
  committed
  d8834a1
- feat: added feature for using local models in the dialogue_baseline
  michaelharlam
  committed
  3798e9c
Commits on Jan 28, 2026
- fix: memory storage
  michaelharlam
  committed
  3936d9f
- fix: by ruff
  michaelharlam
  committed
  8963da4
Commits on Jan 29, 2026
- fix: removed comments & hardcode
  michaelharlam
  committed
  e38c3d4
- feat: added docs to the main classes and methods
  michaelharlam
  committed
  0b395ed
Commits on Jan 31, 2026
- feat: project restructured, added makefile (now it doesn't work), prompt is divided, all of system messages are assembled into the one
  michaelharlam
  committed
  93f50fe
- fix: refactored
  michaelharlam
  committed
  04fca3d
Commits on Feb 1, 2026
- fix: removed vestige prompt
  michaelharlam
  committed
  ba33cbc
- fix: removed vestige prompt
  michaelharlam
  committed
  8bba95f
Commits on Feb 4, 2026
- feat: refactoring, added more tests
  michaelharlam
  committed
  99c0474
Commits on Feb 10, 2026
- feat: added semantic similarity for f1
  michaelharlam
  committed
  e05c4b4
Commits on Feb 11, 2026
- feat: added calculating by logs, args similarities, the stub od the llm-as-a-judge
  michaelharlam
  committed
  88dd410
- feat: llm as a judge template is updated
  michaelharlam
  committed
  8d194f0
Commits on Feb 14, 2026
- feat: llm as a judge refreshed
  michaelharlam
  committed
  dd37671
Commits on Feb 15, 2026
- feat: added counting tokens to the langchain messages
  michaelharlam
  committed
  d7f4366
Commits on Apr 27, 2026
- fix: make ci pass
  michaelharlam
  committed
  2522964

Commits

Commits on Nov 9, 2025

Commits on Nov 12, 2025

Commits on Nov 15, 2025

Commits on Nov 16, 2025

Commits on Nov 17, 2025

Commits on Nov 18, 2025

Commits on Dec 25, 2025

Commits on Dec 27, 2025

Commits on Dec 28, 2025

Commits on Jan 6, 2026

Commits on Jan 7, 2026

Commits on Jan 24, 2026

Commits on Jan 28, 2026

Commits on Jan 29, 2026

Commits on Jan 31, 2026

Commits on Feb 1, 2026

Commits on Feb 4, 2026

Commits on Feb 10, 2026

Commits on Feb 11, 2026

Commits on Feb 14, 2026

Commits on Feb 15, 2026

Commits on Apr 27, 2026