mirror of
https://github.com/hwchase17/langchain.git
synced 2025-09-07 05:52:15 +00:00
9129318466
# Causal program-aided language (CPAL) chain ## Motivation This builds on the recent [PAL](https://arxiv.org/abs/2211.10435) to stop LLM hallucination. The problem with the [PAL](https://arxiv.org/abs/2211.10435) approach is that it hallucinates on a math problem with a nested chain of dependence. The innovation here is that this new CPAL approach includes causal structure to fix hallucination. For example, using the below word problem, PAL answers with 5, and CPAL answers with 13. "Tim buys the same number of pets as Cindy and Boris." "Cindy buys the same number of pets as Bill plus Bob." "Boris buys the same number of pets as Ben plus Beth." "Bill buys the same number of pets as Obama." "Bob buys the same number of pets as Obama." "Ben buys the same number of pets as Obama." "Beth buys the same number of pets as Obama." "If Obama buys one pet, how many pets total does everyone buy?" The CPAL chain represents the causal structure of the above narrative as a causal graph or DAG, which it can also plot, as shown below.  . The two major sections below are: 1. Technical overview 2. Future application Also see [this jupyter notebook](https://github.com/borisdev/langchain/blob/master/docs/extras/modules/chains/additional/cpal.ipynb) doc. ## 1. Technical overview ### CPAL versus PAL Like [PAL](https://arxiv.org/abs/2211.10435), CPAL intends to reduce large language model (LLM) hallucination. The CPAL chain is different from the PAL chain for a couple of reasons. * CPAL adds a causal structure (or DAG) to link entity actions (or math expressions). * The CPAL math expressions are modeling a chain of cause and effect relations, which can be intervened upon, whereas for the PAL chain math expressions are projected math identities. PAL's generated python code is wrong. It hallucinates when complexity increases. ```python def solution(): """Tim buys the same number of pets as Cindy and Boris.Cindy buys the same number of pets as Bill plus Bob.Boris buys the same number of pets as Ben plus Beth.Bill buys the same number of pets as Obama.Bob buys the same number of pets as Obama.Ben buys the same number of pets as Obama.Beth buys the same number of pets as Obama.If Obama buys one pet, how many pets total does everyone buy?""" obama_pets = 1 tim_pets = obama_pets cindy_pets = obama_pets + obama_pets boris_pets = obama_pets + obama_pets total_pets = tim_pets + cindy_pets + boris_pets result = total_pets return result # math result is 5 ``` CPAL's generated python code is correct. ```python story outcome data name code value depends_on 0 obama pass 1.0 [] 1 bill bill.value = obama.value 1.0 [obama] 2 bob bob.value = obama.value 1.0 [obama] 3 ben ben.value = obama.value 1.0 [obama] 4 beth beth.value = obama.value 1.0 [obama] 5 cindy cindy.value = bill.value + bob.value 2.0 [bill, bob] 6 boris boris.value = ben.value + beth.value 2.0 [ben, beth] 7 tim tim.value = cindy.value + boris.value 4.0 [cindy, boris] query data { "question": "how many pets total does everyone buy?", "expression": "SELECT SUM(value) FROM df", "llm_error_msg": "" } # query result is 13 ``` Based on the comments below, CPAL's intended location in the library is `experimental/chains/cpal` and PAL's location is`chains/pal`. ### CPAL vs Graph QA Both the CPAL chain and the Graph QA chain extract entity-action-entity relations into a DAG. The CPAL chain is different from the Graph QA chain for a few reasons. * Graph QA does not connect entities to math expressions * Graph QA does not associate actions in a sequence of dependence. * Graph QA does not decompose the narrative into these three parts: 1. Story plot or causal model 4. Hypothetical question 5. Hypothetical condition ### Evaluation Preliminary evaluation on simple math word problems shows that this CPAL chain generates less hallucination than the PAL chain on answering questions about a causal narrative. Two examples are in [this jupyter notebook](https://github.com/borisdev/langchain/blob/master/docs/extras/modules/chains/additional/cpal.ipynb) doc. ## 2. Future application ### "Describe as Narrative, Test as Code" The thesis here is that the Describe as Narrative, Test as Code approach allows you to represent a causal mental model both as code and as a narrative, giving you the best of both worlds. #### Why describe a causal mental mode as a narrative? The narrative form is quick. At a consensus building meeting, people use narratives to persuade others of their causal mental model, aka. plan. You can share, version control and index a narrative. #### Why test a causal mental model as a code? Code is testable, complex narratives are not. Though fast, narratives are problematic as their complexity increases. The problem is LLMs and humans are prone to hallucination when predicting the outcomes of a narrative. The cost of building a consensus around the validity of a narrative outcome grows as its narrative complexity increases. Code does not require tribal knowledge or social power to validate. Code is composable, complex narratives are not. The answer of one CPAL chain can be the hypothetical conditions of another CPAL Chain. For stochastic simulations, a composable plan can be integrated with the [DoWhy library](https://github.com/py-why/dowhy). Lastly, for the futuristic folk, a composable plan as code allows ordinary community folk to design a plan that can be integrated with a blockchain for funding. An explanation of a dependency planning application is [here.](https://github.com/borisdev/cpal-llm-chain-demo) --- Twitter handle: @boris_dev --------- Co-authored-by: Boris Dev <borisdev@Boriss-MacBook-Air.local>
Readme tests(draft)
Integrations Tests
Prepare
This repository contains functional tests for several search engines and databases. The tests aim to verify the correct behavior of the engines and databases according to their specifications and requirements.
To run some integration tests, such as tests located in
tests/integration_tests/vectorstores/, you will need to install the following
software:
- Docker
- Python 3.8.1 or later
We have optional group test_integration in the pyproject.toml file. This group
should contain dependencies for the integration tests and can be installed using the
command:
poetry install --with test_integration
Any new dependencies should be added by running:
# add package and install it after adding:
poetry add tiktoken@latest --group "test_integration" && poetry install --with test_integration
Before running any tests, you should start a specific Docker container that has all the
necessary dependencies installed. For instance, we use the elasticsearch.yml container
for test_elasticsearch.py:
cd tests/integration_tests/vectorstores/docker-compose
docker-compose -f elasticsearch.yml up
Prepare environment variables for local testing:
- copy
tests/.env.exampletotests/.env - set variables in
tests/.envfile, e.gOPENAI_API_KEY
Additionally, it's important to note that some integration tests may require certain
environment variables to be set, such as OPENAI_API_KEY. Be sure to set any required
environment variables before running the tests to ensure they run correctly.
Recording HTTP interactions with pytest-vcr
Some of the integration tests in this repository involve making HTTP requests to external services. To prevent these requests from being made every time the tests are run, we use pytest-vcr to record and replay HTTP interactions.
When running tests in a CI/CD pipeline, you may not want to modify the existing cassettes. You can use the --vcr-record=none command-line option to disable recording new cassettes. Here's an example:
pytest --log-cli-level=10 tests/integration_tests/vectorstores/test_pinecone.py --vcr-record=none
pytest tests/integration_tests/vectorstores/test_elasticsearch.py --vcr-record=none
Run some tests with coverage:
pytest tests/integration_tests/vectorstores/test_elasticsearch.py --cov=langchain --cov-report=html
start "" htmlcov/index.html || open htmlcov/index.html