Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Qwen-AgentWorld: Language World Models for General Agents

arxiv.org

Post reply on HN
Print view View original

51–60 of 65 posts

Re: Qwen-AgentWorld: Language World Models for General Agents

#51

Post by roenxi »

androiddrew wrote: post #41

I understand what the model is doing. I am struggling to understand where this is going to fit in a workflow. I understand a big gap is that any LLM based ai agent isn't aware of the consequences of its actions because it barely understands the future state its actions will have, hence this model that can. So, is this like a bolt on where you have an agent powered by an LLM, then the world model reviews the action it…

> I understand a big gap is that any LLM based ai agent isn't aware of the consequences of its actions because it barely understands the future state its actions will have, hence this model that can.

These are probably equivalent. Ie, awareness of consequences is the same as understanding the future state. And the present state for that matter, I don't see how someone could be said to understand something if they can't predict the consequences of interacting with it. It is forcing the model to develop a more complex internal world model.

Re: Qwen-AgentWorld: Language World Models for General Agents

#52

Post by jvanderbot »

kakugawa wrote: post #22
dmos62 wrote: post #20

Earlier quoted context omitted.

So, if I'm reading this correctly, whereas a regular LLM would, given a prompt to edit a file, infer a sed call, this "world" model infers the resulting contents of the file.

Here's the demo: https://docs.qwenlm.ai/resources/mlu56_demo.html Here's the description of the world model prompt for the web domain: "A precise GUI state simulator — given the current screen (as HTML) and a user action, predicts the exact next screen as a complete, self-contained HTML document." (You can click the world model prompt box to expand it and see the full prompt.) So the world model generates the current…

And a world model is useful for ... action space search which would require prediction?

Re: Qwen-AgentWorld: Language World Models for General Agents

#53

Post by verdverm »

khimaros wrote: post #5
verdverm wrote: post #2

35B model from the qwen-3.5 line https://github.com/QwenLM/Qwen-AgentWorld https://huggingface.co/Qwen/Qwen-AgentWorld-35B-A3B

unsloth, activate!

I'm using official @8bit quants from Qwen, they maintain more capability

Re: Qwen-AgentWorld: Language World Models for General Agents

#54

Post by khimaros »

npodbielski wrote: post #35
adrian_b wrote: post #17

The smaller of the two models is open weights and available on Huggingface: https://huggingface.co/Qwen/Qwen-AgentWorld-35B-A3B

I tried to run it but seems like it is either broken or it does not work on dockerized llama.cpp: 0.01.865.326 E llama_model_load: error loading model: missing tensor 'blk.40.attn_norm.weight'

that particular quant is just corrupted. these work but seem to loop in reasoning a lot https://huggingface.co/groxaxo/Qwen-AgentWorld-35B-A3B-GGUF

Re: Qwen-AgentWorld: Language World Models for General Agents

#55

Post by tsunamifury »

This questions the nature of banning SOTA models like fable deeply.

As simpler models with better simulated context will be able to more practically execute than SOTAs without such training.

To me this says we should open fable up for defensive reasons rather than fear offensive use. SOTA models will be continuously outmatched by better technique lower grade models with better context techniques like this plus longer walks and deeper inference.

Now you might says SOTAs then could use that and go even further… but how are you going to keep that cat in the bag anyways?

Re: Qwen-AgentWorld: Language World Models for General Agents

#56

Post by anana_ »

androiddrew wrote: post #41

I understand what the model is doing. I am struggling to understand where this is going to fit in a workflow. I understand a big gap is that any LLM based ai agent isn't aware of the consequences of its actions because it barely understands the future state its actions will have, hence this model that can. So, is this like a bolt on where you have an agent powered by an LLM, then the world model reviews the action it…

It looks like the purpose of this model is to i. generate environmental sim data for doing RL on other models or ii. act as a foundation model (they trained it to select actions as well as predicting the next state in the same loop?)

Either way, neither are intended for end consumers.

Re: Qwen-AgentWorld: Language World Models for General Agents

#59

Post by dippogriff »

nostrebored wrote: post #28
dippogriff wrote: post #9

I'm a fan of this direction. For me the most interesting use case for these world models isn't even training, it's verification. If this thing or some idealized version of it can actually reliably simulate state transitions, could you use it to verify an agent's execution path against hard constraints and replace/eclipse LLMs-as-a-judge?

Well if you can do this then you don't delegate execution path derivation to the agent. The benefit is a predictable coherent world state where you understand the impact of { current state } x { action } without having to enumerate that huge cartesian product.

[dead]
Post reply on HN