Tech experts are starting to doubt that ChatGPT and A.I. ‘hallucinations’ will ever go away: ‘This isn’t fixable’::Experts are starting to doubt it, and even OpenAI CEO Sam Altman is a bit stumped.

  • Zeth0s@lemmy.world
    link
    fedilink
    English
    arrow-up
    42
    arrow-down
    1
    ·
    edit-2
    1 year ago

    That’s called context. For chatgpt it is a bit less than 4k words. Using api it goes up to a bit less of 32k. Alternative models goes up to a bit less than 64k.

    Model wouldn’t know anything you said before that

    That is one of the biggest limitations of current generation of LLMs.

    • Womble@lemmy.world
      link
      fedilink
      English
      arrow-up
      3
      ·
      1 year ago

      Thats not 100% true. they also work by modifying meanings of words based on context and then those modified meanings propagate indefinitely forwards. But yes, direct context is limited so things outside it arent directly used.

      • Zeth0s@lemmy.world
        link
        fedilink
        English
        arrow-up
        1
        ·
        1 year ago

        They don’t really chance the meaning of the words, they just look for the “best” words given the recent context, by taking into account the different possible meanings of the words

        • Womble@lemmy.world
          link
          fedilink
          English
          arrow-up
          1
          ·
          edit-2
          1 year ago

          No they do, thats one of the key innovations of LLMs the attention and feed forward steps where they propagate information from related words into each other based on context. from https://www.understandingai.org/p/large-language-models-explained-with?r=cfv1p

          For example, in the previous section we showed a hypothetical transformer figuring out that in the partial sentence “John wants his bank to cash the,” his refers to John. Here’s what that might look like under the hood. The query vector for his might effectively say “I’m seeking: a noun describing a male person.” The key vector for John might effectively say “I am: a noun describing a male person.” The network would detect that these two vectors match and move information about the vector for John into the vector for his.

          • Zeth0s@lemmy.world
            link
            fedilink
            English
            arrow-up
            1
            ·
            edit-2
            1 year ago

            That’s exactly what I said

            They don’t really chance the meaning of the words, they just look for the “best” words given the recent context, by taking into account the different possible meanings of the words

            The word’s meanings haven’t changed, but the model can choose based on the context accounting for the different meanings of words

            • Womble@lemmy.world
              link
              fedilink
              English
              arrow-up
              1
              ·
              1 year ago

              The key vector for John might effectively say “I am: a noun describing a male person.” The network would detect that these two vectors match and move information about the vector for John into the vector for his.

              This is the bit you are missing, the attention network actively changes the token vectors depending on context, this is transferring new information into the meanings of that word.

              • Zeth0s@lemmy.world
                link
                fedilink
                English
                arrow-up
                1
                ·
                edit-2
                1 year ago

                The network doesn’t detect matches, but the model definitely works on similarities. Words are mapped in a hyperspace, with the idea that that space can mathematically retain conceptual similarity as spatial representation.

                Words are transformed in a mathematical representation that is able (or at least tries) to retain semantic information of words.

                But different meanings of the different words belongs to the words themselves and are defined by the language, model cannot modify them.

                Anyway we are talking about details here. We could kill the audience of boredom

                Edit. I asked gpt-4 to summarize the concepts. I believe it did a decent job. I hope it helps:

                1. Embedding Space:

                  • Initially, every token is mapped to a point (or vector) in a high-dimensional space via embeddings. This space is typically called the “embedding space.”
                  • The dimensionality of this space is determined by the size of the embeddings. For many Transformer models, this is often several hundred dimensions, e.g., 768 for some versions of GPT and BERT.
                2. Positional Encodings:

                  • These are vectors added to the embeddings to provide positional context. They share the same dimensionality as the embedding vectors, so they exist within the same high-dimensional space.
                3. Transformations Through Layers:

                  • As tokens’ representations (vectors) pass through Transformer layers, they undergo a series of linear and non-linear transformations. These include matrix multiplications, additions, and the application of functions like softmax.
                  • At each layer, the vectors are “moved” within this high-dimensional space. When we say “moved,” we mean they are transformed, resulting in a change in their coordinates in the vector space.
                  • The self-attention mechanism allows a token’s representation to be influenced by other tokens’ representations, effectively “pulling” or “pushing” it in various directions in the space based on the context.
                4. Nature of the Vector Space:

                  • This space is abstract and high-dimensional, making it hard to visualize directly. However, in this space, the “distance” and “direction” between vectors can have semantic meaning. Vectors close to each other can be seen as semantically similar or related.
                  • The exact nature and structure of this space are learned during training. The model adjusts the parameters (like weights in the attention mechanisms and feed-forward networks) to ensure that semantically or syntactically related concepts are positioned appropriately relative to each other in this space.
                5. Output Space:

                  • The final layer of the model transforms the token representations into an output space corresponding to the vocabulary size. This is a probability distribution over all possible tokens for the next word prediction.

                In essence, the entire process of token representation within the Transformer model can be seen as continuous transformations within a vector space. The space itself can be considered a learned representation where relative positions and directions hold semantic and syntactic significance. The model’s training process essentially shapes this space in a way that facilitates accurate and coherent language understanding and generation.

                • Womble@lemmy.world
                  link
                  fedilink
                  English
                  arrow-up
                  1
                  ·
                  1 year ago

                  Yes of course it works on similarities, I havent disputed that. My point was that the transformations of the token vectors are a transfer of information, and that this transfer of information is not lost as things move out of the context length. That information may slowly decohere over time if it is not reinforced, but the model does not immediately forget things as they move out of context as you originally claimed.

                  • Zeth0s@lemmy.world
                    link
                    fedilink
                    English
                    arrow-up
                    1
                    ·
                    edit-2
                    1 year ago

                    It does, as model only works with a well defined chunk of tokens of a given length. Everything before is lost. Clearly part of the information of previous context is in that chunk.

                    But let’s say that I am talking about wine, at some point I talk about chianti. I and the chatbot go on discussing for over 4k words (I am using chatgpt as an example) without mentioning chianti. After that the chatbot will know we are discussing about wine, but it won’t know we covered the topic of chianti.

                    This is what I meant.