Developmental machine learning studies how artificial agents can model the way children learn open-ended repertoires of skills. Such agents need to create and represent goals, select which ones to pursue and learn to achieve them. Recent approaches have considered goal spaces that were either fixed and hand-defined or learned using generative models of states. This limited agents to sample goals within the distribution of known effects. We argue that the ability to imagine out-of-distribution goals is key to enable creative discoveries and open-ended learning.

The recent advent of reasoning models like OpenAI's o1 was met with excited speculation by the AI community about the mechanisms underlying these capabilities in closed models, followed by a rush of replication efforts, particularly from the open source community. These speculations were largely settled by the demonstration from DeepSeek-R1 that chains-of-thought and reinforcement learning (RL) can effectively replicate reasoning on top of base LLMs. However, it remains valuable to explore alternative methods for theoretically eliciting reasoning that could help elucidate the underlying mechanisms, as well as providing additional methods that may offer complementary benefits. Here, we build on the long-standing literature in cognitive psychology and cognitive architectures, which postulates that reasoning arises from the orchestrated, sequential execution of a set of modular, predetermined cognitive operations. Crucially, we implement this key idea within a modern agentic tool-calling framework. In particular, we endow an LLM with a small set of "cognitive tools" encapsulating specific reasoning operations, each executed by the LLM itself. Surprisingly, this simple strategy results in considerable gains in performance on standard mathematical reasoning benchmarks compared to base LLMs, for both closed and open-weight models. For instance, providing our "cognitive tools" to GPT-4.1 increases its pass@1 performance on AIME2024 from 32% to 53%, even surpassing the performance of o1-preview. In addition to its practical implications, this demonstration contributes to the debate regarding the role of post-training methods in eliciting reasoning in LLMs versus the role of inherent capabilities acquired during pre-training, and whether post-training merely uncovers these latent abilities.

Additional Feedback: Line-by-line comments: Figure 1 - This figure is quite cluttered. I would recommend removing/simplifying some of the graphics (e.g. the thought bubbles) and, when possible, moving them outside the environment canvas. Line 37 - Do the citations on lines 38-39 substantiate the preceding claim that language actually influences children's exploration behavior? This seems like a very difficult claim to test, how do you disentangle mental maturity with language acquisition? Note that I do not consider children "narrating their ongoing activities" to be a meaningful change in behavior if it is not accompanied by a change in how the children actually complete those activities.

Reviewers agreed that the paper proposes an interesting model for learning language conditioned goal reaching policies and performs a thorough investigation on a simple task. There was agreement that the environment studied in the paper is quite simplistic and that the paper would benefit from a task with a richer grammar/goal space. Nevertheless, the results on the task in the paper are sufficiently interesting for acceptance as a poster.

Developmental machine learning studies how artificial agents can model the way children learn open-ended repertoires of skills. Such agents need to create and represent goals, select which ones to pursue and learn to achieve them. Recent approaches have considered goal spaces that were either fixed and hand-defined or learned using generative models of states. This limited agents to sample goals within the distribution of known effects. We argue that the ability to imagine out-of-distribution goals is key to enable creative discoveries and open-ended learning.

Autonomous reinforcement learning agents must be intrinsically motivated to explore their environment, discover potential goals, represent them and learn how to achieve them. As children do the same, they benefit from exposure to language, using it to formulate goals and imagine new ones as they learn their meaning. In our proposed learning architecture (IMAGINE), the agent freely explores its environment and turns natural language descriptions of interesting interactions from a social partner into potential goals. IMAGINE learns to represent goals by jointly learning a language model and a goal-conditioned reward function. Just like humans, our agent uses language compositionality to generate new goals by composing known ones. Leveraging modular model architectures based on Deep Sets and gated-attention mechanisms, IMAGINE autonomously builds a repertoire of behaviors and shows good zero-shot generalization properties for various types of generalization. When imagining its own goals, the agent leverages zero-shot generalization of the reward function to further train on imagined goals and refine its behavior. We present experiments in a simulated domain where the agent interacts with procedurally generated scenes containing objects of various types and colors, discovers goals, imagines others and learns to achieve them.

For years we have been relying on either BPM or RPA as per need to achieve maximum efficiency and value from the business processes. Which ultimately gives an edge to the organisation business. These are still the most relevant tools to improve our processes, but with availability of more technologies and tools it is being critical for businesses now to have an amalgamation of old procedures with new tools, technologies and approaches. Intelligent Process Automation (IPA) refers to the application of Artificial Intelligence and related new technologies, including Computer Vision, Cognitive automation and Machine Learning to Robotic Process Automation. This convergence of technologies produces automation capabilities that dramatically elevate business value and competitive advantages.

While cognitive computing, often referred to as artificial intelligence (AI), is hardly new, the recent level of interest in it is astounding. The combination of vendor marketing, concerns about job losses, and even discussion of "robot overlords" have prompted massive interest in the media. There is also plenty of substance behind the hype. Cognitive technologies offer the possibility of increased productivity, better knowledge-based interactions with customers, and the ability to solve problems that are too complex for human brains. While there have been several "AI winters" and "AI springs" over the past 50 years, there is reason to be confident that the flowering this AI spring is changing the garden permanently.

Machine learning was undoubtedly one of the hottest topics in radiology last year, with a steady stream of academic research papers highlighting how machine learning, particularly deep learning, can outperform traditional algorithms or manual processes in certain use-cases. Investment in machine learning start-ups also continued, with several companies attracting early stage funding. To date, more than $100m has been invested in start-ups that are developing AI solutions for radiology. Furthermore, commercial activity gained pace, with at least 20 companies exhibiting AI-based products at the RSNA conference towards the end of the year, although most were prototypes and only a handful had regulatory clearance. Whilst the enthusiasm for machine learning is certainly justified, it inevitably raises expectations, potentially to unrealistic levels.

Recently I was consulting with a publishing company that is exploring various ways to digitize and contextualize its content. Knowing that some of the company's competitors had signed deals with IBM's Watson, I asked several executives why they had not done a Watson deal themselves. "We think that the market for AI software is rapidly commoditizing, and we believe we can assemble the needed capabilities ourselves at much lower cost," was this company's party line. Some particularly knowledgeable managers mentioned that they expected the company would instead make use of open-source cognitive software made available from various providers. These potential providers are not small vendors -- they include, for example, Google, Facebook, Microsoft, Amazon, and Yahoo.