Repository files navigation

PaperQA2 is a package for doing high-accuracy retrieval augmented generation (RAG) on PDFs or text files,with a focus on the scientific literature.See ourrecent 2024 paper to see examples of PaperQA2's superhuman performance in scientific tasks likequestion answering, summarization, and contradiction detection.

• Quickstart
  • Example Output
• What is PaperQA2
  • PaperQA2 vs PaperQA
  • What's New in Version 5 (aka PaperQA2)?
  • PaperQA2 Algorithm
• Installation
• CLI Usage
  • Bundled Settings
  • Rate Limits
• Library Usage
  • ask manually
  • Adding Documents Manually
  • Async
  • Choosing Model
    • Locally Hosted
  • Changing Embedding Model
    • Local Embedding Models (Sentence Transformers)
  • Adjusting number of sources
  • Using Code or HTML
  • Using External DB/Vector DB and Caching
  • Creating Index
    • Manifest Files
  • Reusing Index
  • Running on LitQA v2
  • Using Clients Directly
• Where do I get papers?
  • Zotero
  • Paper Scraper
• Callbacks
  • Caching Embeddings
• Customizing Prompts
  • Pre and Post Prompts
• FAQ
  • How come I get different results than your papers?
  • How is this different from LlamaIndex or LangChain?
  • Can I save or load?
• Reproduction
• Citation

In this example we take a folder of research paper PDFs,magically get their metadata - including citation counts with a retraction check,then parse and cache PDFs into a full-text search index,and finally answer the user question with an LLM agent.

pip install paper-qacd my_paperspqa ask'How can carbon nanotubes be manufactured at a large scale?'

Question: Has anyone designed neural networks that compute with proteins or DNA?

The claim that neural networks have been designed to compute with DNA is supported by multiple sources. The work by Qian, Winfree, and Bruck demonstrates the use of DNA strand displacement cascades to construct neural network components, such as artificial neurons and associative memories, using a DNA-based system (Qian2011Neural pages 1-2, Qian2011Neural pages 15-16, Qian2011Neural pages 54-56). This research includes the implementation of a 3-bit XOR gate and a four-neuron Hopfield associative memory, showcasing the potential of DNA for neural network computation. Additionally, the application of deep learning techniques to genomics, which involves computing with DNA sequences, is well-documented. Studies have applied convolutional neural networks (CNNs) to predict genomic features such as transcription factor binding and DNA accessibility (Eraslan2019Deep pages 4-5, Eraslan2019Deep pages 5-6). These models leverage DNA sequences as input data, effectively using neural networks to compute with DNA. While the provided excerpts do not explicitly mention protein-based neural network computation, they do highlight the use of neural networks in tasks related to protein sequences, such as predicting DNA-protein binding (Zeng2016Convolutional pages 1-2). However, the primary focus remains on DNA-based computation.

PaperQA2 is engineered to be the best agentic RAG model for working with scientific papers.Here are some features:

• A simple interface to get good answers with grounded responses containing in-text citations.
• State-of-the-art implementation including document metadata-awarenessin embeddings and LLM-based re-ranking and contextual summarization (RCS).
• Support for agentic RAG, where a language agent can iteratively refine queries and answers.
• Automatic redundant fetching of paper metadata,including citation and journal quality data from multiple providers.
• A usable full-text search engine for a local repository of PDF/text files.
• A robust interface for customization, with default support for allLiteLLM models.

By default, it usesOpenAI embeddings andmodels with a Numpy vector DB to embed and search documents. However, you can easily use other closed-source, open-source models or embeddings (see details below).

PaperQA2 depends on some awesome libraries/APIs that make our repo possible.Here are some in no particular order:

1. Semantic Scholar
2. Crossref
3. Unpaywall
4. Pydantic
5. tantivy
6. LiteLLM
7. pybtex
8. PyMuPDF

We've been working on hard on fundamental upgrades for a while and mostly followedSemVer.meaning we've incremented the major version number on each breaking change.This brings us to the current major version number v5.So why call is the repo now called PaperQA2?We wanted to remark on the fact though that we've exceeded human performance onmany important metrics.So we arbitrarily call version 5 and onward PaperQA2,and versions before it as PaperQA1 to denote the significant change in performance.We recognize that we are challenged at naming and counting at FutureHouse,so we reserve the right at any time to arbitrarily change the name to PaperCrow.

Version 5 added:

• A CLIpqa
• Agentic workflows invoking tools forpaper search, gathering evidence, and generating an answer
• Removed much of the statefulness from theDocs object
• A migration to LiteLLM for compatibility with many LLM providersas well as centralized rate limits and cost tracking
• A bundled set of configurations (readhere))containing known-good hyperparameters

Note thatDocs objects pickled from prior versions ofPaperQA are incompatible with version 5,and will need to be rebuilt.Also, our minimum Python version was increased to Python 3.11.

To understand PaperQA2, let's start with the pieces of the underlying algorithm.The default workflow of PaperQA2 is as follows:

The tools can be invoked in any order by a language agent.For example, an LLM agent might do a narrow and broad search,or using different phrasing for the gather evidence step from the generate answer step.

For a non-development setup,install PaperQA2 (aka version 5) fromPyPI.Note version 5 requires Python 3.11+.

pip install paper-qa>=5

For development setup,please refer to theCONTRIBUTING.md file.

PaperQA2 uses an LLM to operate,so you'll need to either set an appropriateAPI key environment variable (i.e.export OPENAI_API_KEY=sk-...)or set up an open source LLM server (i.e. usingllamafile.Any LiteLLM compatible model can be configured to use with PaperQA2.

If you need to index a large set of papers (100+),you will likely want an API key for bothCrossref andSemantic Scholar,which will allow you to avoid hitting public rate limits using these metadata services.Those can be exported asCROSSREF_API_KEY andSEMANTIC_SCHOLAR_API_KEY variables.

The fastest way to test PaperQA2 is via the CLI. First navigate to a directory with some papers and use thepqa cli:

$ pqa ask'What manufacturing challenges are unique to bispecific antibodies?'

You will see PaperQA2 index your local PDF files, gathering the necessary metadata for each of them (usingCrossref andSemantic Scholar),search over that index, then break the files into chunked evidence contexts, rank them, and ultimately generate an answer. The next time this directory is queried, your index will already be built (save for any differences detected, like new added papers), so it will skip the indexing and chunking steps.

All prior answers will be indexed and stored, you can view them by querying via thesearch subcommand, or access them yourself in yourPQA_HOME directory, which defaults to~/.pqa/.

$ pqa search -i'answers''antibodies'

PaperQA2 is highly configurable, when running from the command line,pqa --help shows all options and short descriptions. For example to run with a higher temperature:

$ pqa --temperature 0.5 ask'What manufacturing challenges are unique to bispecific antibodies?'

You can view all settings withpqa view. Another useful thing is to change to other templated settings - for examplefast is a setting that answers more quickly and you can see it withpqa -s fast view

Maybe you have some new settings you want to save? You can do that with

pqa -s my_new_settings --temperature 0.5 --llm foo-bar-5 save

and then you can use it with

pqa -s my_new_settings ask'What manufacturing challenges are unique to bispecific antibodies?'

If you runpqa with a command which requires a new indexing, say if you change the default chunk_size, a new index will automatically be created for you.

pqa --parsing.chunk_size 5000 ask'What manufacturing challenges are unique to bispecific antibodies?'

You can also usepqa to do full-text search with use of LLMs view the search command. For example, let's save the index from a directory and give it a name:

pqa -i nanomaterials index

Now I can search for papers about thermoelectrics:

pqa -i nanomaterials search thermoelectrics

or I can use the normal ask

pqa -i nanomaterials ask'Are there nm scale features in thermoelectric materials?'

Both the CLI and module have pre-configured settings based on prior performance and our publications, they can be invoked as follows:

pqa --settings<setting name> ask'Are there nm scale features in thermoelectric materials?'

Insidepaperqa/configs we bundle known useful settings:

If you are hitting rate limits, say with the OpenAI Tier 1 plan, you can add them into PaperQA2.For each OpenAI tier, a pre-built setting exists to limit usage.

pqa --settings'tier1_limits' ask'Are there nm scale features in thermoelectric materials?'

This will limit your system to use thetier1_limits,and slow down your queries to accommodate.

You can also specify them manually with any rate limit string that matches the specification in thelimits module:

pqa --summary_llm_config'{"rate_limit": {"gpt-4o-2024-08-06": "30000 per 1 minute"}}' ask'Are there nm scale features in thermoelectric materials?'

Or by adding into aSettings object, if calling imperatively:

frompaperqaimportSettings,askanswer=ask("What manufacturing challenges are unique to bispecific antibodies?",settings=Settings(llm_config={"rate_limit": {"gpt-4o-2024-08-06":"30000 per 1 minute"}},summary_llm_config={"rate_limit": {"gpt-4o-2024-08-06":"30000 per 1 minute"}},    ),)

PaperQA2's full workflow can be accessed via Python directly:

frompaperqaimportSettings,askanswer=ask("What manufacturing challenges are unique to bispecific antibodies?",settings=Settings(temperature=0.5,paper_directory="my_papers"),)

Please see ourinstallation docs for how to install the package from PyPI.

The answer object has the following attributes:formatted_answer,answer (answer alone),question , andcontext (the summaries of passages found for answer).ask will use theSearchPapers tool, which will query a local index of files, you can specify this location via theSettings object:

frompaperqaimportSettings,askanswer=ask("What manufacturing challenges are unique to bispecific antibodies?",settings=Settings(temperature=0.5,paper_directory="my_papers"),)

ask is just a convenience wrapper around the real entrypoint, which can be accessed if you'd like to run concurrent asynchronous workloads:

frompaperqaimportSettings,agent_query,QueryRequestanswer=awaitagent_query(QueryRequest(query="What manufacturing challenges are unique to bispecific antibodies?",settings=Settings(temperature=0.5,paper_directory="my_papers"),    ))

The default agent will use an LLM based agent,but you can also specify a"fake" agent to use a hard coded call path of search -> gather evidence -> answer to reduce token usage.

If you prefer fine grained control, and you wish to add objects to the docs object yourself (rather than using the search tool), then the previously existingDocs object interface can be used:

frompaperqaimportDocs,Settings# valid extensions include .pdf, .txt, and .htmldoc_paths= ("myfile.pdf","myotherfile.pdf")docs=Docs()fordocindoc_paths:docs.add(doc)settings=Settings()settings.llm="claude-3-5-sonnet-20240620"settings.answer.answer_max_sources=3answer=docs.query("What manufacturing challenges are unique to bispecific antibodies?",settings=settings,)print(answer.formatted_answer)

PaperQA2 is written to be used asynchronously.The synchronous API is just a wrapper around the async.Here are the methods and theirasync equivalents:

The synchronous version just calls the async version in a loop.Most modern python environments supportasync natively (including Jupyter notebooks!).So you can do this in a Jupyter Notebook:

importasynciofrompaperqaimportDocsasyncdefmain()->None:docs=Docs()# valid extensions include .pdf, .txt, and .htmlfordocin ("myfile.pdf","myotherfile.pdf"):awaitdocs.aadd(doc)answer=awaitdocs.aquery("What manufacturing challenges are unique to bispecific antibodies?"    )print(answer.formatted_answer)asyncio.run(main())

By default, it uses OpenAI models withgpt-4o-2024-08-06 for both the re-ranking and summary step, thesummary_llm setting, and for the answering step, thellm setting. You can adjust this easily:

frompaperqaimportSettings,askanswer=ask("What manufacturing challenges are unique to bispecific antibodies?",settings=Settings(llm="gpt-4o-mini",summary_llm="gpt-4o-mini",paper_directory="my_papers"    ),)

You can use Anthropic or any other model supported bylitellm:

frompaperqaimportSettings,askanswer=ask("What manufacturing challenges are unique to bispecific antibodies?",settings=Settings(llm="claude-3-5-sonnet-20240620",summary_llm="claude-3-5-sonnet-20240620"    ),)

You can use llama.cpp to be the LLM. Note that you should be using relatively large models, because PaperQA2 requires following a lot of instructions. You won't get good performance with 7B models.

The easiest way to get set-up is to download allama file and execute it with-cb -np 4 -a my-llm-model --embedding which will enable continuous batching and embeddings.

frompaperqaimportSettings,asklocal_llm_config=dict(model_list=[dict(model_name="my_llm_model",litellm_params=dict(model="my-llm-model",api_base="http://localhost:8080/v1",api_key="sk-no-key-required",temperature=0.1,frequency_penalty=1.5,max_tokens=512,            ),        )    ])answer=ask("What manufacturing challenges are unique to bispecific antibodies?",settings=Settings(llm="my-llm-model",llm_config=local_llm_config,summary_llm="my-llm-model",summary_llm_config=local_llm_config,    ),)

Models hosted withollama are also supported.To run the example below make sure you have downloaded llama3.2 and mxbai-embed-large via ollama.

frompaperqaimportSettings,asklocal_llm_config= {"model_list": [        {"model_name":"ollama/llama3.2","litellm_params": {"model":"ollama/llama3.2","api_base":"http://localhost:11434",            },        }    ]}answer=ask("What manufacturing challenges are unique to bispecific antibodies?",settings=Settings(llm="ollama/llama3.2",llm_config=local_llm_config,summary_llm="ollama/llama3.2",summary_llm_config=local_llm_config,embedding="ollama/mxbai-embed-large",    ),)

PaperQA2 defaults to using OpenAI (text-embedding-3-small) embeddings, but has flexible options for both vector stores and embedding choices. The simplest way to change an embedding is via theembedding argument to theSettings object constructor:

frompaperqaimportSettings,askanswer=ask("What manufacturing challenges are unique to bispecific antibodies?",settings=Settings(embedding="text-embedding-3-large"),)

embedding accepts any embedding model name supported by litellm. PaperQA2 also supports an embedding input of"hybrid-<model_name>" i.e."hybrid-text-embedding-3-small" to use a hybrid sparse keyword (based on a token modulo embedding) and dense vector embedding, where any litellm model can be used in the dense model name."sparse" can be used to use a sparse keyword embedding only.

Embedding models are used to create PaperQA2's index of the full-text embedding vectors (texts_index argument). The embedding model can be specified as a setting when you are adding new papers to theDocs object:

frompaperqaimportDocs,Settingsdocs=Docs()fordocin ("myfile.pdf","myotherfile.pdf"):docs.add(doc,settings=Settings(embedding="text-embedding-large-3"))

Note that PaperQA2 uses Numpy as a dense vector store.Its design of using a keyword search initially reduces the number of chunks needed for each answer to a relatively small number < 1k.Therefore,NumpyVectorStore is a good place to start, it's a simple in-memory store, without an index.However, if a larger-than-memory vector store is needed, we are currently lacking here.

The hybrid embeddings can be customized:

frompaperqaimport (Docs,HybridEmbeddingModel,SparseEmbeddingModel,LiteLLMEmbeddingModel,)model=HybridEmbeddingModel(models=[LiteLLMEmbeddingModel(),SparseEmbeddingModel(ndim=1024)])docs=Docs()fordocin ("myfile.pdf","myotherfile.pdf"):docs.add(doc,embedding_model=model)

The sparse embedding (keyword) models default to having 256 dimensions, but this can be specified via thendim argument.

You can use aSentenceTransformerEmbeddingModel model if you installsentence-transformers, which isa local embedding library with support for HuggingFace models and more. You can install it by adding thelocal extras.

pip install paper-qa[local]

and then prefix embedding model names withst-:

frompaperqaimportSettings,askanswer=ask("What manufacturing challenges are unique to bispecific antibodies?",settings=Settings(embedding="st-multi-qa-MiniLM-L6-cos-v1"),)

or with a hybrid model

frompaperqaimportSettings,askanswer=ask("What manufacturing challenges are unique to bispecific antibodies?",settings=Settings(embedding="hybrid-st-multi-qa-MiniLM-L6-cos-v1"),)

You can adjust the numbers of sources (passages of text) to reduce token usage or add more context.k refers to the top k most relevant and diverse (may from different sources) passages. Each passage is sent to the LLM to summarize, or determine if it is irrelevant. After this step, a limit ofmax_sources is applied so that the final answer can fit into the LLM context window. Thus,k >max_sources andmax_sources is the number of sources used in the final answer.

frompaperqaimportSettingssettings=Settings()settings.answer.answer_max_sources=3settings.answer.k=5docs.query("What manufacturing challenges are unique to bispecific antibodies?",settings=settings,)

You do not need to use papers -- you can use code or raw HTML. Note that this tool is focused on answering questions, so it won't do well at writing code. One note is that the tool cannot infer citations from code, so you will need to provide them yourself.

importglobimportosfrompaperqaimportDocssource_files=glob.glob("**/*.js")docs=Docs()forfinsource_files:# this assumes the file names are unique in codedocs.add(f,citation="File "+os.path.name(f),docname=os.path.name(f))answer=docs.query("Where is the search bar in the header defined?")print(answer)

You may want to cache parsed texts and embeddings in an external database or file. You can then build a Docs object from those directly:

frompaperqaimportDocs,Doc,Textdocs=Docs()for ...inmy_docs:doc=Doc(docname=...,citation=...,dockey=...,citation=...)texts= [Text(text=...,name=...,doc=doc)for ...inmy_texts]docs.add_texts(texts,doc)

Indexes will be placed in thehome directory by default.This can be controlled via thePQA_HOME environment variable.

Indexes are made by reading files in theSettings.paper_directory.By default, we recursively read from subdirectories of the paper directory,unless disabled usingSettings.index_recursively.The paper directory is not modified in any way, it's just read from.

The indexing process attempts to infer paper metadata like title and DOIusing LLM-powered text processing.You can avoid this point of uncertainty using a "manifest" file,which is a CSV containing three columns (order doesn't matter):

• file_location: relative path to the paper's PDF within the index directory
• doi: DOI of the paper
• title: title of the paper

By providing this information,we ensure queries to metadata providers like Crossref are accurate.

The local search indexes are built based on a hash of the currentSettings object.So make sure you properly specify thepaper_directory to yourSettings object.In general, it's advisable to:

1. Pre-build an index given a folder of papers (can take several minutes)
2. Reuse the index to perform many queries

importosfrompaperqaimportSettingsfrompaperqa.agents.mainimportagent_queryfrompaperqa.agents.modelsimportQueryRequestfrompaperqa.agents.searchimportget_directory_indexasyncdefamain(folder_of_papers:str|os.PathLike)->None:settings=Settings(paper_directory=folder_of_papers)# 1. Build the index. Note an index name is autogenerated when unspecifiedbuilt_index=awaitget_directory_index(settings=settings)print(settings.get_index_name())# Display the autogenerated index nameprint(awaitbuilt_index.index_files)# Display the index contents# 2. Use the settings as many times as you want with askanswer_response_1=awaitagent_query(query=QueryRequest(query="What is the best way to make a vaccine?",settings=settings        )    )answer_response_2=awaitagent_query(query=QueryRequest(query="What manufacturing challenges are unique to bispecific antibodies?",settings=settings,        )    )

Inpaperqa/agents/task.py, you will find:

1. GradablePaperQAEnvironment: an environment that can grade answers given an evaluation function.
2. LitQAv2TaskDataset: a task dataset designed to pull LitQA v2 from Hugging Face,and create oneGradablePaperQAEnvironment per question

Here is an example of how to use them:

importosfromaviary.envimportTaskDatasetfromldp.agentimportSimpleAgentfromldp.alg.callbacksimportMeanMetricsCallbackfromldp.alg.runnersimportEvaluator,EvaluatorConfigfrompaperqaimportQueryRequest,Settingsfrompaperqa.agents.taskimportTASK_DATASET_NAMEasyncdefevaluate(folder_of_litqa_v2_papers:str|os.PathLike)->None:base_query=QueryRequest(settings=Settings(paper_directory=folder_of_litqa_v2_papers)    )dataset=TaskDataset.from_name(TASK_DATASET_NAME,base_query=base_query)metrics_callback=MeanMetricsCallback(eval_dataset=dataset)evaluator=Evaluator(config=EvaluatorConfig(batch_size=3),agent=SimpleAgent(),dataset=dataset,callbacks=[metrics_callback],    )awaitevaluator.evaluate()print(metrics_callback.eval_means)

One of the most powerful features of PaperQA2 is its ability to combine data from multiple metadata sources. For example,Unpaywall can provide open access status/direct links to PDFs,Crossref can provide bibtex, andSemantic Scholar can provide citation licenses. Here's a short demo of how to do this:

frompaperqa.clientsimportDocMetadataClient,ALL_CLIENTSclient=DocMetadataClient(clients=ALL_CLIENTS)details=awaitclient.query(title="Augmenting language models with chemistry tools")print(details.formatted_citation)# Andres M. Bran, Sam Cox, Oliver Schilter, Carlo Baldassari, Andrew D. White, and Philippe Schwaller.#  Augmenting large language models with chemistry tools. Nature Machine Intelligence,# 6:525-535, May 2024. URL: https://doi.org/10.1038/s42256-024-00832-8,# doi:10.1038/s42256-024-00832-8.# This article has 243 citations and is from a domain leading peer-reviewed journal.print(details.citation_count)# 243print(details.license)# cc-byprint(details.pdf_url)# https://www.nature.com/articles/s42256-024-00832-8.pdf

theclient.query is meant to check for exact matches of title. It's a bit robust (like to casing, missing a word). There are duplicates for titles though - so you can also add authors to disambiguate. Or you can provide a doi directlyclient.query(doi="10.1038/s42256-024-00832-8").

If you're doing this at a large scale, you may not want to useALL_CLIENTS (just omit the argument) and you can specify which specific fields you want to speed up queries. For example:

details=awaitclient.query(title="Augmenting large language models with chemistry tools",authors=["Andres M. Bran","Sam Cox"],fields=["title","doi"],)

will return much faster than the first query and we'll be certain the authors match.

Well that's a really good question! It's probably best to just download PDFs of papers you think will help answer your question and start from there.

It's been a while since we've tested this - so let us know if it runs into issues!

If you useZotero to organize your personal bibliography,you can use thepaperqa.contrib.ZoteroDB to query papers from your library,which relies onpyzotero.

Installpyzotero via thezotero extra for this feature:

pip install paperqa[zotero]

First, note that PaperQA2 parses the PDFs of papers to store in the database,so all relevant papers should have PDFs stored inside your database.You can get Zotero to automatically do this by highlighting the referencesyou wish to retrieve, right clicking, and selecting"Find Available PDFs".You can also manually drag-and-drop PDFs onto each reference.

To download papers, you need to get an API key for your account.

1. Get your library ID, and set it as the environment variableZOTERO_USER_ID.
   • For personal libraries, this ID is givenhere at the part "Your userID for use in API calls is XXXXXX".
   • For group libraries, go to your group pagehttps://www.zotero.org/groups/groupname, and hover over the settings link. The ID is the integer after /groups/. (h/t pyzotero!)
2. Create a new API keyhere and set it as the environment variableZOTERO_API_KEY.
   • The key will need read access to the library.

With this, we can download papers from our library and add them to PaperQA2:

frompaperqaimportDocsfrompaperqa.contribimportZoteroDBdocs=Docs()zotero=ZoteroDB(library_type="user")# "group" if group libraryforiteminzotero.iterate(limit=20):ifitem.num_pages>30:continue# skip long papersdocs.add(item.pdf,docname=item.key)

which will download the first 20 papers in your Zotero database and addthem to theDocs object.

We can also do specific queries of our Zotero library and iterate over the results:

foriteminzotero.iterate(q="large language models",qmode="everything",sort="date",direction="desc",limit=100,):print("Adding",item.title)docs.add(item.pdf,docname=item.key)

You can read more about the search syntax by typingzotero.iterate? in IPython.

If you want to search for papers outside of your own collection, I've found an unrelated project calledpaper-scraper that lookslike it might help. But beware, this project looks like it uses some scraping tools that may violate publisher's rights or be in a gray area of legality.

frompaperqaimportDocskeyword_search="bispecific antibody manufacture"papers=paperscraper.search_papers(keyword_search)docs=Docs()forpath,datainpapers.items():try:docs.add(path)exceptValueErrorase:# sometimes this happens if PDFs aren't downloaded or readableprint("Could not read",path,e)answer=docs.query("What manufacturing challenges are unique to bispecific antibodies?")print(answer)

To execute a function on each chunk of LLM completions, you need to provide a function that can be executed on each chunk. For example, to get a typewriter view of the completions, you can do:

deftypewriter(chunk:str)->None:print(chunk,end="")docs=Docs()# add some docs...docs.query("What manufacturing challenges are unique to bispecific antibodies?",callbacks=[typewriter],)

In general, embeddings are cached when you pickle aDocs regardless of what vector store you use. So as long as you save your underlyingDocs object, you should be able to avoid re-embedding your documents.

You can customize any of the prompts using settings.

frompaperqaimportDocs,Settingsmy_qa_prompt= ("Answer the question '{question}'\n""Use the context below if helpful. ""You can cite the context using the key like (Example2012). ""If there is insufficient context, write a poem ""about how you cannot answer.\n\n""Context: {context}")docs=Docs()settings=Settings()settings.prompts.qa=my_qa_promptdocs.query("Are covid-19 vaccines effective?",settings=settings)

Following the syntax above, you can also include prompts thatare executed after the query and before the query. For example, you can use this to critique the answer.

Internally at FutureHouse, we have a slightly different set of tools. We're trying to get some of them, like citation traversal, into this repo. However, we have APIs and licenses to access research papers that we cannot share openly. Similarly, in our research papers' results we do not start with the known relevant PDFs. Our agent has to identify them using keyword search over all papers, rather than just a subset. We're gradually aligning these two versions of PaperQA, but until there is an open-source way to freely access papers (even just open source papers) you will need to provide PDFs yourself.

LangChainandLlamaIndexare both frameworks for working with LLM applications,with abstractions made for agentic workflows and retrieval augmented generation.

Over time, the PaperQA team over time chose to become framework-agnostic,instead outsourcing LLM drivers toLiteLLMand no framework besides Pydantic for its tools.PaperQA focuses on scientific papers and their metadata.

PaperQA can be reimplemented using either LlamaIndex or LangChain.For example, ourGatherEvidence tool can be reimplementedas a retriever with an LLM-based re-ranking and contextual summary.There is similar work with the tree response method in LlamaIndex.

TheDocs class can be pickled and unpickled. This is useful if you want to save the embeddings of the documents and then load them later.

importpickle# savewithopen("my_docs.pkl","wb")asf:pickle.dump(docs,f)# loadwithopen("my_docs.pkl","rb")asf:docs=pickle.load(f)

Contained indocs/2024-10-16_litqa2-splits.json5are the question IDs(correspond withLAB-Bench's LitQA2 question IDs)used in the train and evaluation splits,as well as paper DOIs used to build the train and evaluation splits' indexes.The test split remains held out.

Please read and cite the following papers if you use this software:

@article{skarlinski2024language,title ={Language agents achieve superhuman synthesis of scientific knowledge},author ={    Michael D. Skarlinski and Sam Cox and Jon M. Laurent and James D. Braza and Michaela Hinks and Michael J. Hammerling and Manvitha Ponnapati and Samuel G. Rodriques and Andrew D. White},year ={2024},journal ={arXiv preprent arXiv:2409.13740},url ={https://doi.org/10.48550/arXiv.2409.13740}}

@article{lala2023paperqa,title ={PaperQA: Retrieval-Augmented Generative Agent for Scientific Research},author ={    Jakub Lála and Odhran O'Donoghue and Aleksandar Shtedritski and Sam Cox and Samuel G. Rodriques and Andrew D. White},journal ={arXiv preprint arXiv:2312.07559},year ={2023}}