This weekly roundup thread is intended for all culture war posts. 'Culture war' is vaguely defined, but it basically means controversial issues that fall along set tribal lines. Arguments over culture war issues generate a lot of heat and little light, and few deeply entrenched people ever change their minds. This thread is for voicing opinions and analyzing the state of the discussion while trying to optimize for light over heat.
Optimistically, we think that engaging with people you disagree with is worth your time, and so is being nice! Pessimistically, there are many dynamics that can lead discussions on Culture War topics to become unproductive. There's a human tendency to divide along tribal lines, praising your ingroup and vilifying your outgroup - and if you think you find it easy to criticize your ingroup, then it may be that your outgroup is not who you think it is. Extremists with opposing positions can feed off each other, highlighting each other's worst points to justify their own angry rhetoric, which becomes in turn a new example of bad behavior for the other side to highlight.
We would like to avoid these negative dynamics. Accordingly, we ask that you do not use this thread for waging the Culture War. Examples of waging the Culture War:
-
Shaming.
-
Attempting to 'build consensus' or enforce ideological conformity.
-
Making sweeping generalizations to vilify a group you dislike.
-
Recruiting for a cause.
-
Posting links that could be summarized as 'Boo outgroup!' Basically, if your content is 'Can you believe what Those People did this week?' then you should either refrain from posting, or do some very patient work to contextualize and/or steel-man the relevant viewpoint.
In general, you should argue to understand, not to win. This thread is not territory to be claimed by one group or another; indeed, the aim is to have many different viewpoints represented here. Thus, we also ask that you follow some guidelines:
-
Speak plainly. Avoid sarcasm and mockery. When disagreeing with someone, state your objections explicitly.
-
Be as precise and charitable as you can. Don't paraphrase unflatteringly.
-
Don't imply that someone said something they did not say, even if you think it follows from what they said.
-
Write like everyone is reading and you want them to be included in the discussion.
On an ad hoc basis, the mods will try to compile a list of the best posts/comments from the previous week, posted in Quality Contribution threads and archived at /r/TheThread. You may nominate a comment for this list by clicking on 'report' at the bottom of the post and typing 'Actually a quality contribution' as the report reason.
Jump in the discussion.
No email address required.
Notes -
Sorry, but no. The main effort into multimodal models has been to bolt on multimodal features to a text-trained base model, leading to the absolutely dismal state of vision models. It merely involves, chopping up images and other media into patches, and projecting those into the token embedding space (which is different than tokenizing them), and finetuning an existing model on that information.
Take a look at the LLaVA paper, which while somewhat dated is largely the technique still used on the state of the art for multimodal models.
For a more recent paper, see Qwen 2.5 vision which is also a text-only LLM with vision slapped on top.
Most telling is the fact that none of the top commercially available chatbots have any native capability whatsoever to output images, and just blindly ram your prompt into a diffusion model api. They'll happily generate for you something totally unlike the prompt, and cheerfully insist that it's exactly what you asked for.
TTS is of course fundamentally a sequence task, which maps neatly into an extension of generating text. Bolting on an output head and giving a nice massage of finetuning will straightforwardly give good results. (note that this is fundamentally different from using a separate TTS engine, but also fundamentally different from having a native multimodal model.)
My understanding is that LLaVa has long been supersede by things like cogVLM. I'm not clear on the finer implementation details.
For models like Gemini and the latest GPTs, we have very little public information about their architecture.
GPT-4V was demoed to have image generation capabilities that blew dedicated image models out of the water. OAI hasn't released it, despite strong clamoring, but Altman has said it's on the cards.
The issue you're describing is just poor implementation of image gen, at the very least GPT-4V does astonishingly better.
More options
Context Copy link
More options
Context Copy link