flandre

I ran a quick experiment examining how DeepSeek-R1 carries out on agentic tasks, despite not supporting tool usage natively, and I was rather amazed by initial outcomes. This experiment runs DeepSeek-R1 in a single-agent setup, where the model not only prepares the actions however also develops the actions as executable Python code. On a subset1 of the GAIA recognition split, DeepSeek-R1 outshines Claude 3.5 Sonnet by 12.5% outright, from 53.1% to 65.6% correct, and other models by an even larger margin:

The experiment followed model usage guidelines from the DeepSeek-R1 paper and the design card: Don't utilize few-shot examples, prevent adding a system prompt, and set the temperature level to 0.5 - 0.7 (0.6 was used). You can find more evaluation details here.

Approach

DeepSeek-R1's strong coding abilities enable it to serve as an agent without being clearly trained for tool use. By enabling the model to generate actions as Python code, townshipmarket.co.za it can flexibly communicate with environments through code execution.

Tools are carried out as Python code that is consisted of straight in the timely. This can be a simple function definition or a module of a bigger package - any valid Python code. The design then creates code actions that call these tools.

Results from executing these actions feed back to the design as follow-up messages, driving the next actions up until a last response is reached. The agent structure is an easy iterative coding loop that mediates the discussion between the model and its environment.

Conversations

DeepSeek-R1 is used as chat model in my experiment, where the model autonomously pulls extra context from its environment by utilizing tools e.g. by using a search engine or bring information from web pages. This drives the discussion with the environment that continues up until a last answer is reached.

On the other hand, o1 are known to perform improperly when utilized as chat designs i.e. they do not try to pull context during a conversation. According to the connected post, o1 designs perform best when they have the full context available, with clear guidelines on what to do with it.

Initially, I likewise attempted a full context in a single prompt method at each step (with results from previous actions included), however this resulted in substantially lower ratings on the GAIA subset. Switching to the conversational approach explained above, I was able to reach the reported 65.6% performance.

This raises an intriguing concern about the claim that o1 isn't a chat model - maybe this observation was more relevant to older o1 models that did not have tool usage abilities? After all, isn't tool usage support a crucial mechanism for making it possible for designs to pull extra context from their environment? This conversational technique certainly seems effective for DeepSeek-R1, though I still need to carry out similar experiments with o1 designs.

Generalization

Although DeepSeek-R1 was mainly trained with RL on mathematics and yewiki.org coding tasks, it is exceptional that generalization to agentic jobs with tool use by means of code actions works so well. This capability to generalize to agentic tasks reminds of recent research by DeepMind that shows that RL generalizes whereas SFT remembers, although generalization to tool use wasn't investigated in that work.

Despite its capability to generalize to tool use, DeepSeek-R1 often produces long thinking traces at each action, compared to other designs in my experiments, restricting the effectiveness of this model in a single-agent setup. Even simpler tasks often take a very long time to complete. Further RL on agentic tool usage, be it by means of code actions or not, could be one alternative to improve effectiveness.

Underthinking

I likewise observed the underthinking phenomon with DeepSeek-R1. This is when a reasoning design frequently changes between various reasoning thoughts without adequately exploring appealing courses to reach a right service. This was a significant reason for excessively long reasoning traces produced by DeepSeek-R1. This can be seen in the recorded traces that are available for download.

Future experiments

Another common application of thinking designs is to use them for planning just, while utilizing other designs for wavedream.wiki producing code actions. This could be a prospective brand-new feature of freeact, if this separation of functions shows beneficial for more complex jobs.

I'm likewise curious about how reasoning designs that already support tool usage (like o1, o3, ...) perform in a single-agent setup, with and without creating code actions. Recent developments like OpenAI's Deep Research or Hugging Face's open-source Deep Research, which also uses code actions, look interesting.