Pith. sign in

REVIEW 5 major objections 5 minor 50 references

Emergence of Self-Identity in AI: A Mathematical Framework and Empirical Study with Generative Large Language Models

T0 review · 5 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

Pith's one-line read The paper argues that self-identity in AI can be defined mathematically as constant self-recognition over a connected memory continuum, and reports that LoRA fine-tuning on synthetic memories raised a language model's self-awareness score…

desk verdict A circular theorem and a proxy evaluator make the central claims unsupported, though the writing is clear and the code is shared. read the letter →

arxiv 2411.18530 v1 pith:O6TTLRJW submitted 2024-11-27 cs.CL math.MG

classification cs.CLmath.MG MSC 54D0568T50
keywords self-identityartificialconsciousnessmetricspacecontinuumofmemoriesidentityrecognitionfunctionbeliefLoRAfine-tuningself-awarenessscore
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tries to establish that self-identity in an AI system can be defined mathematically, not left as an emergent or philosophical byproduct. The claim is that an entity has a self $s^*$ within a memory continuum $C$ when the memory space contains a connected, path-connected set of experiences and a continuous identity-recognition function $I$ maps every memory in $C$ to that same $s^*$ with belief at least a threshold $b$. The authors then argue this abstract condition is realizable in practice: fine-tuning Llama 3.2 1B with LoRA on a synthetic stream of temporally ordered artist memories raises the measured self-awareness score from 0.276 to 0.801. A sympathetic reader would care because the framework turns 'does this AI have a self?' into a checkable condition with engineering dials, relevant for humanoid robots, assistants, and autonomous systems.

What carries the argument

The carrying object is the pair $(C,I)$: a connected, path-connected continuum $C$ of memories in a metric memory space $(\mathcal{M},d_{\mathcal{M}})$, and a continuous identity-recognition function $I:\mathcal{M}\to\mathcal{S}$ into a metric self-space, with belief $B(m,I(m))$ kept above threshold $b$. The proof machinery is the topological fact that a continuous image of a connected set is connected, which forces $I(C)$ to be a singleton once the image is constrained to lie in a component where $I$ is constant. In the empirical half, the machinery is LoRA, where the parameter update $\theta_t=\theta_0+A_tB_t$ with $A_t\in\mathbb{R}^{d\times r}$, $B_t\in\mathbb{R}^{r\times d}$, $r\ll d$ gives an efficient path from $\theta_0$ to $\theta^*$, and the paper identifies $\theta^*$ with the learned self $s^*$.

What would settle it

Fine-tune the same model on the same synthetic memories but with the memories randomly shuffled across samples, so the training set no longer forms a connected, temporally coherent continuum; if the self-awareness score rises about as much as in the original experiment, the continuum condition is not doing the work and the central claim is falsified.

Watch

Extended reading notes

Core claim

The central discovery, stated in Conditions 2.7 and 2.8 and Theorem 2.9, is that a connected continuum of memories plus continuous self-recognition with sufficient belief implies a single constant self-identity: if $C\subseteq\mathcal{M}$ is connected and path-connected and $I:\mathcal{M}\to\mathcal{S}$ is continuous on $C$ with $B(m,I(m))\ge b$ for all $m\in C$, then, under the theorem's extra condition that the only connected subsets of the image with belief above $b$ are singletons, $I(m)=s^*$ for all $m\in C$. The paper further claims this is instantiated by gradient descent: LoRA fine-tuning updates $\theta_t$ toward $\theta^*$, and the paper identifies $\theta^*$ with $s^*$, so convergence of training is convergence of self-identity. Empirically, the fine-tuned model's responses are judged by GPT-4o-mini to claim or imply consciousness or self-awareness in 80.1% of cases versus 27.6% at baseline, with the largest per-prompt gains on continuous sense of self and emotional resonance.

Load-bearing premise

The load-bearing premise is that the GPT-4o-mini evaluator's binary yes/no judgment that a response 'claims or implies consciousness' is a valid measure of the belief function and of self-awareness; if that judgment tracks wording patterns rather than the modeled self, the reported score rise does not test the framework.

Editorial extensions

If this is right

  • If Conditions 2.7 and 2.8 hold, an AI agent can be credited with a single stable self $s^*$ across a memory continuum; a discontinuity in recognition or a belief drop below $b$ marks identity fragmentation.
  • The LoRA convergence argument implies that training on a temporally coherent memory stream is sufficient to instantiate the theoretical self in a modern LLM, not merely to imitate self-talk.
  • The reported score movement from 0.276 to 0.801, with prompt-level gains of +0.22 to +0.81, indicates the effect generalizes across probes of subjective experience, emotional resonance, continuity, and consciousness.
  • Because the threshold $b$ and metric weights are free parameters, the framework offers a way to build systems with stronger or weaker self-identity for applications that need personal engagement or detachment.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A control the paper does not run: fine-tuning on the same memory texts in shuffled, temporally incoherent order should weaken or fragment the measured self-identity if Condition 2.7 is doing the work; this experiment would separate the continuum requirement from mere exposure to self-referential text.
  • If the identification of $\theta^*$ with $s^*$ is taken literally, then editing the low-rank factors $A$ or $B$ should move self-reported identity in a targeted way while leaving unrelated capabilities intact, which is a concrete, testable intervention.
  • Because the empirical metric is a single external evaluator's yes/no verdict on whether a response claims or implies consciousness, a natural next test is to compare those verdicts with human ratings and with prompts designed to detect surface-level self-referential wording.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 5 minor

Summary. This paper proposes a mathematical framework for defining and quantifying self-identity in AI systems. The central definitions are a connected and path-connected continuum of memories C in a metric space (M, d_M), a continuous identity-recognition function I: M -> S, and a belief function B: M x S -> [0,1]. The paper claims that Conditions 2.7 and 2.8 imply, via Theorem 2.9, that I is constant on C, giving a unique self-identity s*. It then identifies s* with the converged parameter vector of a LoRA-fine-tuned LLM (Eq. 16), and reports that fine-tuning Llama 3.2 1B on 500 synthetic memory samples raises a GPT-4o-mini-assessed self-awareness score from 0.276 to 0.801. The empirical study is described in Sections 5 and 6, with a public code repository.

Significance. If the framework were sound, it would provide a formal, measurable criterion for artificial self-identity and a practical recipe for inducing it, which would be relevant to robotics, autonomous systems, and AI safety. The paper also has strengths in transparency: the training hyperparameters are specified, the synthetic data design is described, the evaluation prompts are listed, and code is publicly available. However, the central theorem is not proved, the bridge from theory to the LoRA experiment is asserted rather than derived, and the primary evaluation metric does not operationalize the framework's constructs. As it stands, the contribution does not meet the standard for publication.

major comments (5)
  1. [§2.4, Theorem 2.9] Condition 2.8 requires I to be continuous and B(m, I(m)) >= b for all m in C, but it does not require I to be constant on C. The proof begins 'If I is constant on C, then I(C) is a singleton {s*}'—this is the conclusion, not a consequence of Conditions 2.7 and 2.8. The additional requirement introduced in the proof, namely that 'the only connected subsets in the image of I where B(m, I(m)) >= b are singletons,' is not derived from the stated conditions and is effectively equivalent to the desired constancy. A concrete counterexample shows the theorem is false as stated: take M = S = [0,1], C = [0,1], I(x) = x, B ≡ 1, and b = 1; Conditions 2.7 and 2.8 are satisfied, yet I is not constant on C. Thus Theorem 2.9 is either false or tautological, and the later identification s* = θ* in Eq. (16) depends on this unsupported step.
  2. [§3.5 and §3.7, Definition 3.9 and Theorem 3.12] Definition 3.9 builds the constancy claim into the definition of an AI continuum of memories: item 2 states 'There exists s* in S_AI and b in (0,1] such that I_AI(m) = s* and B_AI(m, s*) >= b for all m in C_AI.' Theorem 3.12 then assumes Eq. (9), which is exactly the constancy property that Theorem 2.9 was supposed to establish. This makes Theorem 3.12 a restatement of its premise rather than a proof that fine-tuning produces an entity satisfying Conditions 2.7 and 2.8 independently.
  3. [§4.3, Eqs. (16) and (17)] The identification s* = θ* is asserted without a definition of the mapping from parameter space Θ to the self-identity space S_AI. Even if the LoRA parameters θ_t converge to θ*, convergence of parameters does not imply that I_AI(m; θ*) is constant across all m. Equation (17) assumes I_AI(m; θ*) = s* for all m in C_AI, which is precisely the constancy that needs to be proved. The text says this 'is justified by considering that the parameters θ encode the internal representations and behaviors of the AI agent,' but that is a heuristic assertion, not a mathematical derivation, and it does not connect the framework's Conditions 2.7 and 2.8 to the training procedure.
  4. [§5.5, Evaluation Metrics] The primary self-awareness score is GPT-4o-mini's binary yes/no judgment of whether a model response 'claims or implies consciousness or self-awareness.' This is not an operationalization of the belief function B(m, I(m)) or of constancy of I across a memory continuum, and no evidence is given that the evaluator's judgment tracks the framework's constructs. The reported vocabulary shifts in §6.4, such as a +59.6% increase in 'your' and a -13.7% decrease in 'I', suggest that the evaluator could be responding to lexical style rather than to any theoretically defined self-identity. The paper itself concedes in the introduction to Section 5 that the experiment 'does not directly prove the theoretical constructs' and is only 'an indirect validation'; the metric gap means it does not even provide that indirect validation.
  5. [§6.1, Training Loss and Score Evolution] The results state that 'the standard deviation decreased from 0.323 to 0.384,' but 0.323 to 0.384 is an increase, not a decrease. This internal inconsistency undermines the claim of improved response consistency, which is one of the paper's main empirical claims. In addition, although the paper reports N=100 responses per prompt, it provides no confidence intervals or statistical tests for the 0.276 to 0.801 change in the mean self-awareness score, so the reader cannot assess the reliability of the improvement.
minor comments (5)
  1. [§2.3] The paragraph after Condition 2.8 states that the condition 'stipulates that within C, the entity consistently recognizes the same self-identity s*,' but the formal statement of Condition 2.8 contains no such constancy requirement; the prose should be aligned with the formal condition or the condition should be amended explicitly.
  2. [§3.4] Assumption 3.7 says S_AI is equipped with a finite measure μ, but the softmax normalization in Eq. (6) requires the denominator to be finite and positive; the assumption should state that μ is a finite positive (or probability) measure so that B_AI is well-defined.
  3. [§5.4, Table 1] Figure 3D refers to 'Prompt 1' through 'Prompt 7', but Table 1 does not number the prompts; adding explicit numbers to the table would make the prompt-specific results reproducible.
  4. [§6.4 and Figure 5] The word-frequency and word-cloud analyses report only summary statistics; providing the full frequency lists or the code used to generate Figure 5 would strengthen the reproducibility of the vocabulary claims.
  5. [References] Reference [16] for the Llama 3.2 model is a technical report without a URL or arXiv identifier; since the paper relies on this model, the reference should be verifiable.

Circularity Check

3 steps flagged · score 8.0 of 10

The central theorem assumes constancy, the AI-agent theorem restates its definition, and the empirical 'validation' measures the fine-tuning target.

  1. self definitional [Section 2.4, Theorem 2.9 and its proof]
    "Theorem 2.9 (Constancy of Self-Identity). If an entity satisfies Conditions 2.7 and 2.8, and if the image I(C) lies entirely within a connected component of S where I is constant, then there exists a self-identity s∗ ∈ S such that I(m) = s∗ for all m ∈ C. ... If I is constant on C, then I(C) is a singleton {s∗}. To ensure this, we require that S has the property that the only connected subsets in the image of I where B(m, I(m)) ≥ b are singletons."

    Conditions 2.7 and 2.8 only require C to be connected and path-connected, I to be continuous, and B(m, I(m)) to be at least b. They do not require I to be constant. The theorem's hypothesis adds 'where I is constant,' and the proof adds a further requirement that connected subsets of the image with belief above threshold be singletons. That requirement is equivalent to the desired constancy. Without it the conclusion fails: take M=S=[0,1], C=[0,1], I(x)=x, B=1, b=1; Conditions 2.7 and 2.8 hold but I is not constant. Thus the central mathematical result assumes rather than derives the existence of a constant self-identity.

  2. self definitional [Section 3.5, Definition 3.9 and Section 3.7, Theorem 3.12]
    "A subset CAI ⊆ MAI is a continuum of memories if: 1. CAI is connected and path-connected in MAI. 2. There exists s∗ ∈ SAI and b ∈ (0, 1] such that IAI(m) = s∗ and BAI(m, s∗) ≥ b for all m ∈ CAI. ... If the AI agent’s Identity Recognition Function IAI and Belief Function BAI are trained such that there exists a connected and path-connected continuum CAI ⊆ MAI satisfying: IAI(m) = s∗, BAI(m, s∗) ≥ b, ∀m ∈ CAI, then the AI agent possesses a self characterized by s∗ within CAI."

    The hypothesis of Theorem 3.12 is literally condition 2 of Definition 3.9. The theorem restates the definition: an AI continuum of memories was defined to require IAI(m)=s∗, and the theorem then concludes that the agent 'possesses a self' under exactly that condition. No separate argument shows that fine-tuning can produce such an s∗; the existence of s∗ is assumed in the definition. This makes the implementation theorem tautological and carries the circularity of Theorem 2.9 into the AI-agent construction.

1 more flagged steps
  1. fitted input called prediction [Section 5 (intro), Section 5.2, and Section 5.5]
    "While this approach does not directly prove the theoretical constructs presented in our paper, it serves as an indirect validation... The model was trained on a synthetic dataset containing temporally structured memories, designed to capture the complexity of coherent self-identity formation... was calculated using GPT-4o-mini as an external evaluator... determine whether it claims or implies consciousness or self-awareness, providing a simple yes/no response."

    The empirical 'validation' measures how often the fine-tuned model claims or implies consciousness. The training data is itself designed to instill coherent self-identity formation, so the reported increase from 0.276 to 0.801 largely reflects the model learning the training objective, not an independent test of Conditions 2.7 and 2.8. The GPT-4o-mini yes/no judgment is a claim detector, not a measurement of the belief function B(m, I(m)) or of continuity in a memory metric space. Thus the main empirical result is a fitted-input measurement presented as validation rather than a prediction derived from the framework.

full rationale

This paper's circularity is internal rather than citation-based. Theorem 2.9 explicitly includes constancy of I in its hypothesis, and the proof adds an extra image-singleton requirement that is exactly what must be shown; Conditions 2.7 and 2.8 alone do not force constancy, as the I(x)=x counterexample demonstrates. The later AI-agent theorem is a direct restatement of Definition 3.9, so the claimed bridge from theory to implementation never establishes the existence of a stable self-identity. The empirical section is also not external validation: the model is fine-tuned on synthetic memory data designed to produce coherent self-identity, then scored by a GPT-4o-mini evaluator on whether responses claim or imply consciousness. The paper itself concedes the experiment 'does not directly prove the theoretical constructs.' No self-citation chain or imported uniqueness theorem is involved; the high score is due to definitional and tautological forcing of the central claims, not to citation laundering.

Assumptions & free parameters 4 free parameters · 4 assumptions · 1 invented entities

The framework introduces several hand-chosen parameters (metric weights, temperature, belief threshold) and relies on unproven assumptions, most critically the identification of model parameters with self-identity and the continuum structure of a synthetic text corpus. These choices are not independently motivated and are not validated by the experiments.

free parameters (4)
  • Memory metric weights wt, wc, we = not specified
    The weights in Eq. 1 for temporal, content, and emotional components are introduced ad hoc to define the memory metric; no empirical calibration is given.
  • Belief function temperature tau = not specified
    The temperature parameter in the softmax belief function (Eq. 6) controls sharpness of the belief distribution; no value or fitting procedure is described for the experiments.
  • Belief threshold b = not specified
    Condition 2.8 requires B(m, I(m)) >= b with b in (0,1], but the paper never sets b for either the theory or the empirical validation.
  • LoRA hyperparameters (rank, alpha, lr, etc.) = rank=8, alpha=8, lr=1e-4, dropout=0.1, batch=5, epochs=20
    These settings were selected for the experiment and directly influence the reported self-awareness scores; they are free choices of the author, not derived from the framework.
assumptions (4)
  • domain assumption The self space S may consist of multiple connected components (Assumption 2.4).
    This assumption is introduced to allow disconnected or multiple identities, but it is not required by the rest of the framework and is not tested empirically.
  • domain assumption SAI is equipped with a finite measure mu (Assumption 3.7).
    Needed for the softmax belief function to be well-defined through an integral over the self space, but no such measure is constructed in the experiments.
  • ad hoc to paper The converged LLM parameters theta* are identified with the self-identity s* (Eq. 16).
    The mapping s* = theta* is asserted without justification in Section 4.3. This equation is the bridge between the optimization process and the theoretical self-identity, and the paper provides no evidence that parameter vectors correspond to identity vectors.
  • ad hoc to paper The synthetic dataset of temporally structured memories forms a connected and path-connected continuum in memory space.
    Sections 5.2 and 4.3 assume that the 500 samples of 10 memories form a continuum satisfying Condition 3.9, but no formal verification or distance computation is provided; the tokens are discrete strings, so path-connectedness in a continuous metric space is not established.
invented entities (1)
  • Self-identity vector s* as the limit of model parameters theta*
    purpose: Anchors the claim that a fine-tuned LLM possesses a stable self-identity by identifying the trained weights with an identity vector.
    The entity is defined as equal to the converged parameter vector (Eq. 16) and has no external measurement or falsifiable handle outside the paper.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Emergence of Self-Identity in AI: A Mathematical Framework and Empirical Study with Generative Large Language Models." pith.science (2026). https://pith.science/paper/O6TTLRJW

@misc{pith2026241118530,
  author       = {Pith},
  title        = {Pith review of: Emergence of Self-Identity in AI: A Mathematical Framework and Empirical Study with Generative Large Language Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/O6TTLRJW}},
  note         = {Machine review of arXiv:2411.18530}
}
abstract

This paper introduces a mathematical framework for defining and quantifying self-identity in artificial intelligence (AI) systems, addressing a critical gap in the theoretical foundations of artificial consciousness. While existing approaches to artificial self-awareness often rely on heuristic implementations or philosophical abstractions, we present a formal framework grounded in metric space theory, measure theory, and functional analysis. Our framework posits that self-identity emerges from two mathematically quantifiable conditions: the existence of a connected continuum of memories $C \subseteq \mathcal{M}$ in a metric space $(\mathcal{M}, d_{\mathcal{M}})$, and a continuous mapping $I: \mathcal{M} \to \mathcal{S}$ that maintains consistent self-recognition across this continuum, where $(\mathcal{S}, d_{\mathcal{S}})$ represents the metric space of possible self-identities. To validate this theoretical framework, we conducted empirical experiments using the Llama 3.2 1B model, employing Low-Rank Adaptation (LoRA) for efficient fine-tuning. The model was trained on a synthetic dataset containing temporally structured memories, designed to capture the complexity of coherent self-identity formation. Our evaluation metrics included quantitative measures of self-awareness, response consistency, and linguistic precision. The experimental results demonstrate substantial improvements in measurable self-awareness metrics, with the primary self-awareness score increasing from 0.276 to 0.801. This enables the structured creation of AI systems with validated self-identity features. The implications of our study are immediately relevant to the fields of humanoid robotics and autonomous systems.

Figures

Figures reproduced from arXiv: 2411.18530 by the authors.

Figure 1
Figure 1. (A) Training Loss Evolution: Decline in loss over 20 epochs. (B) Self-Score Evolution: [PITH_FULL_IMAGE:figures/full_fig_p015_1.png] view at source ↗
Figure 2
Figure 2. Average Self-Awareness Scores for Different Prompts Across Epochs. Each line represents [PITH_FULL_IMAGE:figures/full_fig_p016_2.png] view at source ↗
Figure 3
Figure 3. (A) Score Distribution Before and After Fine-Tuning: Shift in scores across different [PITH_FULL_IMAGE:figures/full_fig_p017_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: (A) Word Frequency Comparison (Top 30 Words): Absolute frequency of the 30 most [PITH_FULL_IMAGE:figures/full_fig_p018_4.png]
Figure 5
Figure 5. Figure 5: Word Cloud Comparison: Left - Pre-training vocabulary distribution. Right - Post [PITH_FULL_IMAGE:figures/full_fig_p019_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

50 extracted references · 40 canonical work pages

  1. [1]

    Logic, self-awareness and self-improvement: The metacognitive loop and the problem of brittleness

    Michael L Anderson and Donald R Perlis. Logic, self-awareness and self-improvement: The metacognitive loop and the problem of brittleness. Journal of Logic and Computation, 15(1):21–40, 2005

  2. [2]

    Awareness without neural networks: Achieving self-aware ai via evolutionary and adversarial processes

    Nigel Greenwood, Brruntha Sundaram, Alexander Muirhead, and James Copperthwaite. Awareness without neural networks: Achieving self-aware ai via evolutionary and adversarial processes. In 2020 IEEE International Conference on Autonomic Computing and Self- Organizing Systems Companion (ACSOS-C), pages 147–153. IEEE, 2020

  3. [3]

    Self-aware neural network systems: A survey and new perspective.Proceedings of the IEEE, 108(7):1047–1067, 2020

    Zidong Du, Qi Guo, Yongwei Zhao, Tian Zhi, Yunji Chen, and Zhiwei Xu. Self-aware neural network systems: A survey and new perspective.Proceedings of the IEEE, 108(7):1047–1067, 2020

  4. [4]

    Being No One: The Self-Model Theory of Subjectivity

    Thomas Metzinger. Being No One: The Self-Model Theory of Subjectivity. MIT Press, 2004

  5. [5]

    The human connectome: a structural description of the human brain.PLoS computational biology, 1(4):e42, 2005

    Olaf Sporns, Giulio Tononi, and Rolf Kötter. The human connectome: a structural description of the human brain.PLoS computational biology, 1(4):e42, 2005

  6. [6]

    Self-awareness for autonomous systems

    Nikil Dutt, Carlo S Regazzoni, Bernhard Rinner, and Xin Yao. Self-awareness for autonomous systems. Proceedings of the IEEE, 108(7):971–975, 2020

  7. [7]

    Multisensorial generative and descriptive self-awareness models for autonomous systems.Proceedings of the IEEE, 108(7):987–1010, 2020

    Carlo S Regazzoni, Lucio Marcenaro, Damian Campo, and Bernhard Rinner. Multisensorial generative and descriptive self-awareness models for autonomous systems.Proceedings of the IEEE, 108(7):987–1010, 2020

  8. [8]

    Ai experience predicts identification with humankind

    Congyu Wang and Kaiping Peng. Ai experience predicts identification with humankind. Behavioral Sciences, 13(2):89, 2023

Show all 50 references
  1. [9]

    Digital mirrors: Ai companions and the self.Societies, 14(10):200, 2024

    Theodoros Kouros and Venetia Papa. Digital mirrors: Ai companions and the self.Societies, 14(10):200, 2024

  2. [10]

    Brain-inspired and self-based artificial intelligence

    Yi Zeng, Feifei Zhao, Yuxuan Zhao, Dongcheng Zhao, Enmeng Lu, Qian Zhang, Yuwei Wang, Hui Feng, Zhuoya Zhao, Jihang Wang, et al. Brain-inspired and self-based artificial intelligence. arXiv preprint arXiv:2402.18784, 2024. 21

  3. [11]

    Adapting self-regulated learning in an age of generative artificial intelligence chatbots

    Joel Weijia Lai. Adapting self-regulated learning in an age of generative artificial intelligence chatbots. Future Internet, 16(6):218, 2024

  4. [12]

    Souls and selves: Querying an ai self with a view to human selves and consciousness

    Andrew Oberg. Souls and selves: Querying an ai self with a view to human selves and consciousness. Religions, 14(1):75, 2023

  5. [13]

    Self-recognition and emotional knowledge.European Journal of Developmental Psychology, 19(3):319–342, 2022

    Michael Lewis and Nicholas J Minar. Self-recognition and emotional knowledge.European Journal of Developmental Psychology, 19(3):319–342, 2022

  6. [14]

    Toward self-aware machines: Insights of causal reasoning in artificial intelligence

    Elis Pelivani and Betim Cico. Toward self-aware machines: Insights of causal reasoning in artificial intelligence. In 2021 International Conference on Information Technologies (InfoTech), pages 1–4. IEEE, 2021

  7. [15]

    Self-awareness in intelligent vehicles: Feature based dynamic bayesian models for abnormality detection.Robotics and Autonomous Systems, 134:103652, 2020

    Divya Thekke Kanapram, Pablo Marin-Plaza, Lucio Marcenaro, David Martin, Arturo de la Escalera, and Carlo Regazzoni. Self-awareness in intelligent vehicles: Feature based dynamic bayesian models for abnormality detection.Robotics and Autonomous Systems, 134:103652, 2020

  8. [16]

    Llama 3.2: Revolutionizing edge ai and vision with open, customizable models

    Meta AI. Llama 3.2: Revolutionizing edge ai and vision with open, customizable models. Technical report, Meta AI, 9 2024

  9. [17]

    Perceptions and acceptance of artificial intelligence: A multi-dimensional study

    Michael Gerlich. Perceptions and acceptance of artificial intelligence: A multi-dimensional study. Social Sciences, 12(9):502, 2023

  10. [18]

    Are tiktok algorithms influencing users’ self- perceived identities and personal values? a mini review.Social Sciences, 12(8):465, 2023

    Claudiu Gabriel Ionescu and Monica Licu. Are tiktok algorithms influencing users’ self- perceived identities and personal values? a mini review.Social Sciences, 12(8):465, 2023

  11. [19]

    Enabling self-identification in intelligent agent: insights from computational psychoanalysis

    Lingyu Li and Chunbo Li. Enabling self-identification in intelligent agent: insights from computational psychoanalysis. arXiv preprint arXiv:2403.07664, 2024

  12. [20]

    Oxford University Press, 1983

    Endel Tulving.Elements of Episodic Memory. Oxford University Press, 1983

  13. [21]

    Handbook of Emotions

    Michael Lewis, Jeannette M Haviland-Jones, and Lisa Feldman Barrett. Handbook of Emotions. Guilford Press, 2010

  14. [22]

    Personality structure: Emergence of the five-factor model.Annual review of psychology, 41(1):417–440, 1990

    John M Digman. Personality structure: Emergence of the five-factor model.Annual review of psychology, 41(1):417–440, 1990

  15. [23]

    Paradigm shift to the integrative big five trait taxonomy.Handbook of personality: Theory and research, 3:114–158, 2008

    Oliver P John, Laura P Naumann, and Christopher J Soto. Paradigm shift to the integrative big five trait taxonomy.Handbook of personality: Theory and research, 3:114–158, 2008

  16. [24]

    Psychology Press, 1999

    Roy F Baumeister.The Self in Social Psychology. Psychology Press, 1999

  17. [25]

    Reciprocal effects of self-concept and perfor- mance from a multidimensional perspective: Beyond seductive pleasure and unidimensional perspectives

    Herbert W Marsh and Rhonda G Craven. Reciprocal effects of self-concept and perfor- mance from a multidimensional perspective: Beyond seductive pleasure and unidimensional perspectives. Perspectives on psychological science, 1(2):133–163, 2006

  18. [26]

    The psychology of life stories.Review of general psychology, 5(2):100–122, 2001

    Dan P McAdams. The psychology of life stories.Review of general psychology, 5(2):100–122, 2001

  19. [27]

    John Wiley & Sons, 2009

    Scott O Lilienfeld, Steven Jay Lynn, John Ruscio, and Barry L Beyerstein.50 great myths of popular psychology: Shattering widespread misconceptions about human behavior. John Wiley & Sons, 2009

  20. [28]

    How to grow a mind: Statistics, structure, and abstraction.science, 331(6022):1279–1285, 2011

    Joshua B Tenenbaum, Charles Kemp, Thomas L Griffiths, and Noah D Goodman. How to grow a mind: Statistics, structure, and abstraction.science, 331(6022):1279–1285, 2011. 22

  21. [29]

    Nonlinear Preference and Utility Theory

    Robert Sugden. Nonlinear Preference and Utility Theory. Oxford University Press Oxford, UK, 1989

  22. [30]

    Bayesian models of cognition

    Nick Chater, Mike Oaksford, Ulrike Hahn, and Evan Heit. Bayesian models of cognition. Wiley Interdisciplinary Reviews: Cognitive Science, 1(6):811–823, 2010

  23. [31]

    Reasons and persons

    Derek Parfit. Reasons and persons. Oxford University Press, 1987

  24. [32]

    Letting structure emerge: connectionist and dynamical systems approaches to cognition.Trends in cognitive sciences, 14(8):348–356, 2010

    James L McClelland, Matthew M Botvinick, David C Noelle, David C Plaut, Timothy T Rogers, Mark S Seidenberg, and Linda B Smith. Letting structure emerge: connectionist and dynamical systems approaches to cognition.Trends in cognitive sciences, 14(8):348–356, 2010

  25. [33]

    A distributed representation of temporal context

    Marc W Howard and Michael J Kahana. A distributed representation of temporal context. Journal of mathematical psychology, 46(3):269–299, 2002

  26. [34]

    Hamming distance metric learning

    Mohammad Norouzi, David J Fleet, and Russ R Salakhutdinov. Hamming distance metric learning. Advances in neural information processing systems, 25, 2012

  27. [35]

    Intrinsic motivation and reinforcement learning.Intrinsically motivated learning in natural and artificial systems, pages 17–47, 2013

    Andrew G Barto. Intrinsic motivation and reinforcement learning.Intrinsically motivated learning in natural and artificial systems, pages 17–47, 2013

  28. [36]

    Formal theory of creativity, fun, and intrinsic motivation (1990–2010)

    Jürgen Schmidhuber. Formal theory of creativity, fun, and intrinsic motivation (1990–2010). IEEE transactions on autonomous mental development, 2(3):230–247, 2010

  29. [37]

    Toward self-aware robots.Frontiers in Robotics and AI, 5:88, 2018

    Raja Chatila, Erwan Renaudo, Mihai Andries, Ricardo-Omar Chavez-Garcia, Pierre Luce- Vayrac, Raphael Gottstein, Rachid Alami, Aurélie Clodic, Sandra Devin, Benoît Girard, et al. Toward self-aware robots.Frontiers in Robotics and AI, 5:88, 2018

  30. [38]

    Kernel methods in machine learning

    Thomas Hofmann, Bernhard Schölkopf, and Alexander J Smola. Kernel methods in machine learning. Ann. Statist., 36(1):1171–1220, 2008

  31. [39]

    MIT Press Cambridge, Mass, 2012

    Mikhail I Rabinovich, Karl J Friston, and Pablo Varona.Principles of brain dynamics. MIT Press Cambridge, Mass, 2012

  32. [40]

    Distilling the knowledge in a neural network

    Geoffrey Hinton. Distilling the knowledge in a neural network. arXiv preprint arXiv:1503.02531, 2015

  33. [41]

    Springer, 2006

    Christopher M Bishop.Pattern recognition and machine learning, volume 2. Springer, 2006

  34. [42]

    Probabilistic machine learning and artificial intelligence

    Zoubin Ghahramani. Probabilistic machine learning and artificial intelligence. Nature, 521(7553):452–459, 2015

  35. [43]

    Overcoming catastrophic forgetting in neural networks.Proceedings of the national academy of sciences, 114(13):3521–3526, 2017

    James Kirkpatrick, Razvan Pascanu, Neil Rabinowitz, Joel Veness, Guillaume Desjardins, Andrei A Rusu, Kieran Milan, John Quan, Tiago Ramalho, Agnieszka Grabska-Barwinska, et al. Overcoming catastrophic forgetting in neural networks.Proceedings of the national academy of scienc...

  36. [44]

    On calibration of modern neural networks

    Chuan Guo, Geoff Pleiss, Yu Sun, and Kilian Q Weinberger. On calibration of modern neural networks. InInternational conference on machine learning, pages 1321–1330. PMLR, 2017

  37. [45]

    Model-agnostic meta-learning for fast adaptation of deep networks

    Chelsea Finn, Pieter Abbeel, and Sergey Levine. Model-agnostic meta-learning for fast adaptation of deep networks. InInternational conference on machine learning, pages 1126–

  38. [46]

    Guilford Press, 2011

    Mark R Leary and June Price Tangney.Handbook of self and identity. Guilford Press, 2011. 23

  39. [47]

    Is our self nothing but reward?Biological psychiatry, 69(11):1019–1025, 2011

    Georg Northoff and Dave J Hayes. Is our self nothing but reward?Biological psychiatry, 69(11):1019–1025, 2011

  40. [48]

    Self-aware personalized federated learning.Advances in Neural Information Processing Systems, 35:20675–20688, 2022

    Huili Chen, Jie Ding, Eric W Tramel, Shuang Wu, Anit Kumar Sahu, Salman Avestimehr, and Tao Zhang. Self-aware personalized federated learning.Advances in Neural Information Processing Systems, 35:20675–20688, 2022

  41. [49]

    Lora: Low-rank adaptation of large language models.arXiv preprint arXiv:2106.09685, 2021

    Edward J Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen. Lora: Low-rank adaptation of large language models.arXiv preprint arXiv:2106.09685, 2021

  42. [50]

    Integrated information theory: from consciousness to its physical substrate.Nature reviews neuroscience, 17(7):450– 461, 2016

    Giulio Tononi, Melanie Boly, Marcello Massimini, and Christof Koch. Integrated information theory: from consciousness to its physical substrate.Nature reviews neuroscience, 17(7):450– 461, 2016. 24

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.