worldofideas123

I was referring to the prolog programming language. What would happen if we replace the rationale with a concrete program in prolog: Example tag = daughter(X), rationale = father(_,X),female(X); mother(_,X),female(X).

(1) This is like a new prolog in which the rules are linguistic rules (rationales), and the distilling step-by-step is a way of learning the rules. Call the new prolog lprorat (learning programming with rationales).

(2) Hypothesis: The success of distilling chain of thought learning is that it provides a pseudo large context window. The input pair (label,rationale) allows the student LLM to mimic having a large context window (the context window of a human agent that informs the system by providing the rationale data). So it reduces the embedding distance between words for which entailment is difficult to obtain with short context windows.

(3) The above hypothesis suggests to perform the training process in two phases (a) and (b): (a) A small model is used as a teacher, this allows the learning model to connect d-distant words. (b) A larger LLM is used as a teacher, the training process continues with new data using the rationale provided by a stronger larger LLM.

In phase (a) the learning system learns to infer d-distant words relations, in phase (b) the system is already prepared to relate (2*d)-distant words because it have developed the necessary skills phase (a).

(4) The two-phase approach can be directly generalized to a multi-phase approach using three meta-parameters: the number of teachers, the relative power of each LLM teacher, and the percentage of data in the training set for each teacher.

(5) Need of pruning: From a geometrical point of view, each phase allows the system to strengthen relations between distant words, increasing latent spaces density promotes part of the latent space getting disconnected, that is creating outliers. Pruning should reduce the risk of outliers that promote hallucinations. Outliers are the germ for establishing relation between very distant words but the rationale data perform that function well, so removing outliers don't pause the learning pace.

(6) It is clear that the rationale data enhances the attention mechanism. This suggests modifying the attention mechanism by giving appropriate weights to the attention with respect to the rationale and self attention for the tag.

(7) It would be desirable for the authors to release those small and powerful LLMs. Therefore giving individuals the power to enter into the SOTA realm and allowing us to create new low latency and cost applications with enhanced privacy.

Edited several times for grammar and new ideas.

The aware demostration

I think a way of avoiding hallucinations is using the same LLM with different values of the temperature parameter. Hallucinations, by their own nature, are prone to have great variance, so a change in temperature implies a big change in the story of facts inferred by the LLM, so it seems a main way of fighting hallucinations is checking coherence of the LLM using different values of temperature. So the probability of hallucinations is just the d(story)/d(temperature). This suggest to investigate how the embedding distance of small episodes change with temperature.

it would be interesting to have more information about the LLM that Inflection is using, for example number of parameters.

The Aware Reasoning Project main task is the creation of agents based on LLM to perform a goal using tools, memory, task decomposition and analysis. In this first video the author gives an example of the alpha version of this project. The main insight is that the communication among the agents shows the progress towards the goal. In subsequent videos the optimized use of resources will be shown and how the system is able to continue developing the infrastructure that allow it to attack the main goal in such a way that each agents has a clear function in the system and the information gathered is well propagated through the network of agents, directors and managers. Special care is dedicated to design sub-goals so that their results can be checked easily and when it is possible modified so to align with the main goal.